RU2773318C2 - Large-scale monocellular transcriptome libraries and methods for their production and use - Google Patents

Abstract

FIELD: biotechnology.

SUBSTANCE: options of a method for the production of a sequencing library containing nucleic acids originating from a set of separate nuclei or cells are proposed.

EFFECT: invention provides for the possibility of assessment of transcriptional dynamics of complex biological systems.

21 cl, 42 dwg, 1 tbl, 4 ex

Description

Cross-reference to related applications

[0001] This application claims the benefit of U.S. Provisional Application Reg. No. 62/680259, filed June 4, 2018, and U.S. Provisional Application Reg. No. 62/821678, filed March 21, 2019, the contents of which are included in this description in full.

Application for Government Sponsorship with Federal Funds

[0002] The present invention was developed with government support under Grant No. DP1 HG007811 issued by the National Institutes of Health. The government has certain rights to this invention.

Field of invention

[0003] Embodiments of the present invention relate to nucleic acid sequencing. In particular, variants of the methods and compositions described herein relate to the production of indexed single cell libraries of transcriptomes and their sequence data acquisition.

Prerequisites for the creation of the invention

[0004] Cells go through various functional and molecular states at various stages of their development, such as the development of a multicellular organism and response to various conditions, such as exposure to a therapeutic agent. Characterization of cell state transition pathways or cell fate is useful for understanding cellular pathways, including cell development and molecular response to environmental change. Thus, for example, regulators of developmental defects can be identified, which allows a better understanding of the mechanisms of action of therapeutic agents on cells.

[0005] Combinatorial single cell indexing ("sci") is a technique that uses pooled barcoding to uniquely label nucleic acids that are abundant in single cells or nuclei. However, current single-cell genomic methods lack sufficient throughput and resolution to provide a general understanding of the molecular states and trajectories of the rapidly diversifying and expanding number of cell types that are normally present during the development of a multicellular organism. Modern genomic methods, carried out using single cells, allow only a quick fixation of the state of the cell, and therefore cannot provide information on the dynamics of the cell transition from one state to another, regulated by internal factors (for example, the internal cell cycle program) and external factors (for example, response of a cell to an external stimulus such as a therapeutic agent).

Essence

[0006] The present invention relates to methods for identifying the dynamics of a cell's transition from one state to another by labeling newly synthesized RNA. Whole and newly synthesized RNA transcriptomes are fixed to characterize the dynamics of transcriptomes between time points at the individual cell level. The present invention also relates to methods that are directed to sequencing single cell mRNAs of interest, thereby eliminating current limitations in detecting changes in the amount of any given transcript. In addition, the present invention relates to methods that reduce the rate of cell loss and remove limitations on the effectiveness of reactions, which allows you to profile more individual cells than previously possible.

[0007] In one embodiment of the invention, the method includes obtaining a plurality of nuclei or cells in a first plurality of compartments, where each compartment contains a subset of nuclei or cells, and labeling newly synthesized RNA in subpopulations of cells or nuclei derived from cells. The RNA molecules in each subset of nuclei or cells are processed to obtain indexed nuclei or cells, where such processing includes adding to the RNA nucleic acids present in each subset of nuclei or cells a first compartment-specific index sequence to obtain indexed DNA nucleic acids present in the indexed nuclei or cells, and then combining the indexed nuclei or cells to create the combined indexed nuclei or cells.

[0008] In another embodiment of the invention, the method includes obtaining a plurality of nuclei or cells in a first plurality of compartments, where each compartment contains a subset of nuclei or cells. Each subpopulation is contacted with reverse transcriptase and primer, which hybridize to a predetermined RNA nucleic acid, resulting in the formation of double-stranded DNA nucleic acids with the primer and the corresponding DNA nucleotide sequence from the messenger RNA nucleic acids. DNA molecules in each subset of nuclei or cells are processed to obtain indexed nuclei or cells, where such processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed nucleic acids present in indexed nuclei or cells, and then combining the indexed nuclei or cells to create the combined indexed nuclei or cells.

[0009] In another embodiment of the invention, the method includes obtaining a plurality of nuclei or cells in a first plurality of compartments, where each compartment contains a subset of nuclei or cells. Each subpopulation is contacted with reverse transcriptase and primer, which hybridize to a predetermined RNA nucleic acid, resulting in the formation of double-stranded DNA nucleic acids with the primer and the corresponding DNA nucleotide sequence from the messenger RNA nucleic acids. DNA molecules in each subset of nuclei or cells are processed to obtain indexed nuclei or cells, where such processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed nucleic acids present in indexed nuclei or cells, and then combining the indexed nuclei or cells to create the combined indexed nuclei or cells. The pooled indexed nuclei or cells are separated and then further processed to add a second compartment-specific index to the DNA molecules, pooled, separated and processed again to add a third compartment-specific index to the DNA molecules.

[0010] Definitions

[0011] The terms used here have the generally accepted meaning known to those skilled in the art, unless otherwise noted. Some of the terms used here and their meanings are given below.

[0012] As used herein, the terms "organism" and "individual" are synonymous and refer to microorganisms (eg, prokaryotic or eukaryotic), animals, and plants. An example of an animal is a mammal such as a human.

[0013] As used herein, the term "cell type" refers to the identification of cells based on morphology, phenotype, evolutionary development, or other known or recognizable distinctive properties of cells. Different types of cells can be obtained from the same organism (or from an organism of the same species). Typical cell types include, but are not limited to, gametes (including female gametes, such as mature eggs or ova, and male gametes, such as spermatozoa), ovarian epithelial cells, ovarian fibroblasts, testicular cells, bladder cells, immune cells, B- cells, T cells, natural killer cells, dendritic cells, cancer cells, eukaryotic cells, stem cells, blood cells, muscle cells, fat cells, skin cells, nerve cells, bone cells, pancreatic cells, endothelial cells, epithelial cells pancreas, pancreatic alpha cells, pancreatic beta cells, pancreatic endothelial cells, bone marrow lymphoblasts, bone marrow B lymphoblasts, bone marrow macrophages, bone marrow erythroblasts, bone marrow dendritic cells, bone marrow adipocytes, bone marrow osteocytes , bone marrow chondrocytes, promyeloblasts, bone marrow megakaryoblasts, urinary bladder cells zyra, B-lymphocytes of the brain, glial cells of the brain, neurons, astrocytes of the brain, neuroectoderm cells, macrophages of the brain, microglial cells of the brain, epithelial cells of the brain, neurons of the cortical layer, fibroblasts of the brain, epithelial cells of the mammary gland, epithelial colonic cells, colonic B-lymphocytes, mammary epithelial cells, mammary myoepithelial cells, mammary fibroblasts, colonic enterocytes, cervical epithelial cells, mammary ductal epithelial cells, tongue epithelial cells, tonsil dendritic cells, B-lymphocytes tonsils, peripheral blood lymphoblasts, peripheral blood T lymphoblasts, skin peripheral blood T lymphocytes, peripheral blood natural killer cells, peripheral blood B lymphoblasts, peripheral blood monocytes, peripheral blood myeloblasts, peripheral blood monoblasts, peripheral promyeloblasts blood, peripheral blood macrophages, peripheral blood basophils, liver epithelial cells, liver mast cells, liver epithelial cells, liver B-lymphocytes, spleen endothelial cells, spleen epithelial cells, spleen B-lymphocytes, liver hepatocytes, liver fibroblasts, lung epithelial cells, bronchial endothelial cells, lung fibroblasts, lung B-lymphocytes, lung Schwann cells, lung squamous cells, lung macrophages, lung osteoblasts, neuroendocrine cells, lung alveolar cells, gastric epithelial cells and gastric fibroblasts.

[0014] As used herein, the term "tissue" means a collection or aggregation of cells that together perform one or more specific functions in the body. The cells may be, but need not be, morphologically similar. Typical tissues include, but are not limited to, embryonic tissues, epididymis, eyes, muscles, skin, tendons, veins, arteries, blood, heart, spleen, lymph nodes, bone, bone marrow, lungs, bronchi, trachea, intestines, small intestines, colon, colon, rectum, salivary glands, tongue, gallbladder, appendix, liver, pancreas, brain, stomach, skin, kidney, ureter, bladder, urethra, gonad, testis, ovary, uterus, fallopian tubes, thymus, pituitary, thyroid, adrenal, or parathyroid. The tissue can be obtained from any of the various organs of a human or other organism. The tissue may be healthy or diseased. Examples of diseased tissues include, but are not limited to, malignant neoplasms of the reproductive tissue, lung, breast, colon and rectum, prostate, nasopharynx, stomach, testicles, skin, nervous system, bone, ovary, liver, hematopoietic tissues, pancreas glands, uterus, kidneys, lymphoid tissues, etc. Malignant neoplasms may be neoplasms of various histological subtypes, for example, carcinoma, adenocarcinoma, sarcoma, fibroadenocarcinoma, neuroendocrine tumor, or undifferentiated tumor.

[0015] As used herein, the term "compartment" means an area or volume that separates or isolates some elements from other elements. Typical compartments include, but are not limited to, vials, tubes, wells, drops, boluses, spheres, vessels, surface features, or areas or volumes separated by physical forces such as fluid flow, magnetism, electric current, or the like. In one embodiment of the invention, the compartment is a well of a multi-well plate, such as a 96- or 384-well plate. As used herein, the term "droplets" may include a hydrogel sphere, which is a bead to encapsulate one or more nuclei or cells, and includes a hydrogel composition. In some embodiments, the droplet is a homogeneous droplet of a hydrogel substance or is a hollow droplet having a polymeric hydrogel shell. A homogeneous droplet, whether homogeneous or hollow, may be suitable for encapsulation of one or more nuclei or one or more cells. In some embodiments of the invention, the drop is a drop stabilized with a surfactant.

[0016] As used herein, the term "transposome complex" means an integration enzyme and a nucleic acid comprising an integration recognition site. A "transposome complex" is a functional complex formed by a transposase and a transposase recognition site that is capable of catalyzing a transposition reaction (see, for example, Gunderson et al., WO 2016/130704). Examples of integration enzymes include, but are not limited to, integrase or transposase. Examples of integration recognition sites include, but are not limited to, a transposase recognition site.

[0017] As used herein, the term "nucleic acid" is used in accordance with its use in this field and includes natural nucleic acids or their functional analogues. Particularly suitable functional analogs are capable of hybridizing to a nucleic acid in a sequence-specific manner, or may be used as a template for replication of a particular nucleotide sequence. Natural nucleic acids usually have a backbone containing phosphodiester bonds. The analog structure may have an alternating backbone bond, including any of a variety of bonds known to those skilled in the art. Natural nucleic acids typically have a deoxyribose sugar (eg, present in deoxyribonucleic acid (DNA)) or a ribose sugar (eg, present in ribonucleic acid (RNA)). The nucleic acid may contain any of a number of analogues of these sugar groups known to those skilled in the art. The nucleic acid may include natural or non-natural bases. Accordingly, natural deoxyribonucleic acid may have one or more bases selected from the group consisting of adenine, thymine, cytosine or guanine, and ribonucleic acid may have one or more bases selected from the group consisting of adenine, uracil, cytosine or guanine. Suitable non-natural bases that can be included in the nucleic acid are known to those skilled in the art. Examples of non-natural bases are blocked nucleic acid (LNA), bridged nucleic acid (BNA) and pseudo-complementary bases (Trilink Biotechnologies, San Diego, CA). Bases LNA and BNA can be included in the DNA oligonucleotide, which increase the strength and specificity of hybridization of the oligonucleotides. The bases of LNA and BNA are known to those skilled in the art and are used in routine methods. Unless otherwise noted, the term "nucleic acid" includes natural and non-natural mRNA, non-coding RNA, such as RNA without poly-A at the 3' end, RNA-derived nucleic acids, such as cDNA and DNA.

[0018] As used herein, the term "target", when used in relation to a nucleic acid, means the semantic identifier of the nucleic acid in the methods or compositions described herein and does not necessarily limit the structure or function of the nucleic acid beyond those specifically indicated. The target nucleic acid can be essentially any nucleic acid with a known or unknown sequence. The nucleic acid can be, for example, a fragment of genomic DNA (eg, chromosomal DNA), extrachromosomal DNA such as a plasmid, non-cellular DNA, RNA (eg, RNA or non-coding RNA), proteins (eg, cellular or cell surface proteins), or cDNA. Sequencing allows you to determine the sequence of the entire molecule or part of the target molecule. The target may be derived from a primary nucleic acid sample, such as a nucleus. In one embodiment of the invention, targets can be processed into templates suitable for amplification by attaching universal sequences to one or both ends of each target fragment. Targets can also be obtained from a primary RNA sample by reverse transcription into cDNA. In one embodiment of the invention, the target is used against a subgroup of DNA, RNA or proteins present in the cell. Targeted sequencing involves selecting and isolating genes or regions or proteins of interest, typically by PCR amplification (eg, region-specific primers) or by hybridization-based capture or by using antibodies. Targeted enrichment can be carried out at various stages of this process. Thus, for example, presentation of a target RNA can be achieved using target-specific primers in a reverse transcription step or in a subset hybridization enrichment step from more complex libraries. An example is exome sequencing or L1000 analysis (Subramanian et al., 2017, Cell, 171:1437-1452). Targeted sequencing may include any of the enrichment methods known to those skilled in the art.

[0019] The term "universal", when used here to describe a nucleotide sequence, refers to a region of sequence that is common to two or more nucleic acid molecules, where these molecules also have sequence regions that differ from each other, Universal sequence, which is present in different members of a set of molecules allows the capture of many different nucleic acids due to the group of universal capture nucleic acids, for example, to capture oligonucleotides that are complementary to part of the universal sequence, for example the universal capture sequence. Non-limiting examples of universal capture sequences include sequences that are identical or complementary to primers P5 and P7. Similarly, a universal sequence that is present in various members of a set of molecules allows replication (eg, sequencing) or amplification of many different nucleic acids through a group of universal primers that are complementary to part of the universal sequence, such as the universal anchoring sequence. In one embodiment of the invention, universal anchoring sequences are used as the site to which a universal sequencing primer hybridizes (eg, a sequencing primer for reading in lane 1 or 2). Therefore, the capture oligonucleotide or universal primer includes a sequence that can specifically hybridize to the universal sequence.

[0020] The terms "P5" and "P7" can be used when describing a universal capture sequence or a capture oligonucleotide. The terms "P5" (primer P5) and "P7" (primer P7) refer to the complement to P5 and P7, respectively. It should be noted that any suitable universal capture sequence or capture oligonucleotide may be used in the methods described herein, and that the use of P5 and P7 is exemplary only. The use of capture oligonucleotides such as P5 and P7 or their complements on flow cells is known to those skilled in the art and examples are disclosed in WO 2007/010251, WO 2006/064199, WO 2005/065814, WO 2015/106941, WO 1998 /044151 and WO 2000/018957. Thus, for example, any suitable forward amplification primer, whether immobilized or present in solution, can be used in the complementary sequence hybridization and sequence amplification methods described herein. Likewise, any suitable reverse amplification primer, whether immobilized or present in solution, can be used in the complementary sequence hybridization and sequence amplification methods described herein. One skilled in the art is aware of methods for designing and using primer sequences that are suitable for capturing and/or amplifying the nucleic acids provided herein.

[0021] As used herein, the term "primer" and its derivatives generally refers to any nucleic acid that can hybridize to a target sequence of interest. Typically, the primer functions as a substrate on which nucleotides can be polymerized by the polymerase, or to which a nucleotide sequence, such as an index, can be ligated; however, in some embodiments, the primer may insert into the synthesized nucleic acid strand and form a site to which another primer may hybridize to initiate the synthesis of a new strand that is complementary to the synthesized nucleic acid molecule. The primer may include any combination of nucleotides or their analogues. In some embodiments, the primer is a single stranded oligonucleotide or polynucleotide. As used herein, the terms "polynucleotide" and "oligonucleotide" are synonymous and refer to the polymeric form of nucleotides of any length and may include ribonucleotides, deoxyribonucleotides, their analogs, or mixtures thereof. These terms are to be understood as equivalents, analogs, or any DNA, RNA, cDNA, or antibody-oligonucleotide conjugates derived from nucleotide analogs, and these terms can be applied to single-stranded (such as sense or antisense) polynucleotides and double-stranded polynucleotides. As used herein, the term also encompasses cDNA, which is complementary DNA, or a copy thereof, obtained from messenger RNA, for example by the action of reverse transcriptase. This term refers only to the primary structure of a molecule. Thus, the term includes three-, two- and single-stranded deoxyribonucleic acid ("DNA"), as well as three-, double- and single-stranded ribonucleic acid ("RNA").

[0022] As used herein, the term "adapter" and its derivatives, such as a universal adapter, generally refers to any linear oligonucleotide that can be attached to a nucleic acid molecule according to the invention. In some embodiments of the invention, the adapter is not substantially complementary to the 3' end or 5' end of any target sequence present in the sample. In some embodiments, a suitable adapter is in the range of about 10-100 nucleotides, about 12-60 nucleotides, or about 15-50 nucleotides in length. Generally speaking, the adapter may include any combination of nucleotides and/or nucleic acids. In some aspects of the invention, the adapter may include one or more cleavable groups in one or more positions. In another aspect of the invention, the adapter may include a sequence that is substantially identical to or substantially complementary to at least a portion of a primer, such as a universal primer. In some embodiments, the adapter may include a barcode (also referred to herein as a "tag" or "index") to facilitate subsequent error correction, identification, or sequencing. The terms adapter and adapter are used interchangeably.

[0023] As used herein, the term "each", when used to describe a group of objects, refers to the identification of an individual object in that group, but it does not necessarily refer to every item in that group, unless otherwise noted.

[0024] As used herein, the term "transport" refers to the movement of a molecule through a liquid. This term may include passive transport, such as the movement of molecules along their concentration gradient (eg, passive diffusion). The term can also include active transport, by which molecules can move along a concentration gradient or against a concentration gradient. Thus, transport may involve the application of energy to move one or more molecules in a desired direction or position, such as an amplification site.

[0025] As used herein, the term "amplify", "amplification" or "amplification reaction" and their derivatives, essentially refers to any action or process by which at least a portion of a nucleic acid molecule is replicated or copied into at least one additional nucleic acid molecule. The additional nucleic acid molecule optionally includes a sequence that is substantially identical to or substantially complementary to at least some portion of the template nucleic acid molecule. The template nucleic acid molecule may be single-stranded or double-stranded, and the additional nucleic acid molecule may independently be single-stranded or double-stranded. Amplification includes, but is not limited to, linear or exponential replication of the nucleic acid molecule. In some embodiments of the invention, such amplification can be carried out under isothermal conditions; and in other embodiments of the invention, such amplification may include a thermowell reaction. In some embodiments of the invention, the amplification is multiplex amplification, which includes the simultaneous amplification of multiple target sequences in a single amplification reaction. In some embodiments of the invention, the term "amplification" includes the amplification of at least some of the nucleic acids based on DNA and RNA alone or in combination. The amplification reaction may include any amplification processes known to the person skilled in the art. In some embodiments of the invention, the amplification reaction includes a polymerase chain reaction (PCR).

[0026] As used herein, the term "amplification conditions" and its derivatives generally refers to conditions suitable for amplifying one or more nucleic acid sequences. Such amplification may be linear or exponential. In some embodiments, the amplification conditions may include isothermal conditions, or alternatively, thermowell reaction conditions, or a combination of isothermal conditions and thermowell reaction conditions. In some embodiments of the invention, conditions suitable for amplifying one or more nucleic acid sequences include polymerase chain reaction (PCR) conditions. Typically, amplification conditions refer to a reaction mixture that is sufficient to amplify nucleic acids, such as one or more target sequences flanked by a universal sequence, or to amplify an amplified target sequence ligated to one or more adapters. Generally speaking, amplification conditions include a catalyst for amplification or for nucleic acid synthesis, such as a polymerase; a primer that has some degree of complementarity with the amplifiable nucleic acid; and nucleotides such as deoxyribonucleotide triphosphates (dNTPs) to accelerate primer extension after hybridization with nucleic acid. Amplification conditions may require hybridization or annealing of the primer to the nucleic acid, primer extension, and a denaturation step in which the extended primer is separated from the nucleic acid sequence during amplification. Typically, but not necessarily, amplification conditions may include a thermowell reaction; and in some embodiments of the invention, the amplification conditions include many cycles, where the steps of annealing, extension and separation are repeated. Typically, amplification conditions include cations such as Mg ²⁺ or Mn ²⁺ and may also include various ionic strength modifiers.

[0027] As used herein, the terms "re-amplification" and their derivatives generally refer to any process by which at least a portion of an amplified nucleic acid molecule is further amplified by any suitable amplification reaction (referred to in some embodiments as "secondary" amplification) such that leads to the formation of a re-amplified nucleic acid molecule. Secondary amplification does not need to be identical to the original amplification process that produced the amplified nucleic acid molecule, and it is not required that the re-amplified nucleic acid molecule be completely identical or completely complementary to the amplified nucleic acid molecule; and the only requirement is that the re-amplified nucleic acid molecule includes at least a portion of the amplified nucleic acid molecule or its complement. Thus, for example, re-amplification, as opposed to primary amplification, may involve the use of different amplification conditions and/or different primers, including different target-specific primers.

[0028] As used herein, the term "polymerase chain reaction" ("PCR") refers to the Mullis method, see US Pat. This method of amplifying a polynucleotide of interest consists of introducing a large excess of two oligonucleotide primers into a mixture of DNA containing the polynucleotide of interest, followed by a series of thermal cycles in the presence of a DNA polymerase. The two primers are complementary to the corresponding strands of the double-stranded polynucleotide of interest. The mixture is first denatured at a higher temperature and then the primers are hybridized to complementary sequences in the polynucleotide molecule of interest. After annealing, the primers are extended by polymerase to form a new pair of complementary strands. The steps of denaturation, annealing, and primer extension by polymerase can be repeated many times (and these steps are called thermal cycles) to produce a high concentration of the amplified segment of the desired polynucleotide of interest. The length of the amplified segment of the polynucleotide of interest (amplicon) of interest is determined by the relative positions of the primers to each other, and therefore this length is a controllable parameter. Due to the repetition of such a process, this method is called PCR. Since the desired amplified segments of the polynucleotide of interest become the predominant nucleic acid sequences (in terms of concentration) in the mixture, they are said to be "PCR amplified". In a modification of the method discussed above, target nucleic acid molecules can be amplified by PCR using a variety of different primer pairs, and in some cases, one or more primer pairs per target nucleic acid molecule of interest, allowing for a multiplex PCR reaction. .

[0029] As defined herein, "multiplex amplification" refers to the selective and non-randomized amplification of two or more target sequences in a sample using at least one target-specific primer. In some embodiments of the invention, multiplex amplification is performed such that some or all of the target sequences are amplified in a single reaction vessel. "Plexity" or "multitude" for a given multiplex amplification essentially refers to the number of different target-specific sequences that are amplified during a single multiplex amplification. In some embodiments, the plexity may be approximately 12x, 24x, 48x, 96x, 1692x, 384x, 768x, 1536x, 3072x, 6144x, or more. Amplified target sequences can also be detected by several different methods (for example, by gel electrophoresis followed by densitometry, by quantification using a bioanalyzer or quantitative PCR; by hybridization with a labeled probe; by incorporation of biotinylated primers followed by detection of the "avidin- enzyme"; by incorporating ³² P-labeled deoxynucleotide triphosphates into the amplified target sequence).

[0030] As used herein, "amplified target sequences" and derivatives thereof generally refers to a nucleic acid sequence obtained by amplifying target sequences using target-specific primers and using the methods described herein. Amplified target sequences can be sense (ie, positive strand) or antisense (ie, negative strand) to the target sequences.

[0031] As used herein, the terms "ligation", "ligation" and their derivatives generally refer to a method of covalently linking two or more molecules together, such as covalently linking two or more nucleic acid molecules to each other. In some embodiments of the invention, ligation includes the introduction of "nicknames" between adjacent nucleic acid nucleotides. In some embodiments of the invention, ligation includes the formation of a covalent bond between the end of the first and the end of the second nucleic acid molecule. In some embodiments of the invention, ligation may include the formation of a covalent bond between the 5'-phosphate group of one nucleic acid and the 3'-hydroxyl group of the second nucleic acid, and thereby the formation of a ligated nucleic acid molecule. Typically, for purposes of the disclosure of the present invention, an amplified target sequence may be ligated to an adapter to provide an amplified target sequence ligated to the adapter.

[0032] As used herein, the term "ligase" and its derivatives essentially means any agent capable of catalyzing the ligation of two substrate molecules. In some embodiments, a ligase is an enzyme capable of catalyzing the incorporation of nicks between adjacent nucleic acid nucleotides. In some embodiments, a ligase is an enzyme capable of catalyzing the formation of a covalent bond between the 5'-phosphate of one nucleic acid molecule and the 3'-hydroxyl of another nucleic acid molecule, which will result in the formation of a ligated nucleic acid molecule. Suitable ligases may include, but are not limited to, T4 DNA ligase, T4 RNA ligase, and E. coli DNA ligase.

[0033] As used herein, the term "ligation conditions" and its derivatives generally refers to conditions suitable for ligating two molecules to each other. In some embodiments of the invention, ligation conditions are suitable for closing "nicks" or "gaps" between nucleic acids. The terms "nick" or "gap" used here correspond to the terms used in the literature. Typically, the nick or gap can be ligated in the presence of an enzyme, such as a ligase, at the appropriate temperature and pH. In some embodiments of the invention, T4 DNA ligase can connect "nick" between nucleic acids at a temperature of approximately 70-72°C.

[0034] As used herein, the term "flow cell" refers to a chamber having a solid surface through which one or more liquid reactants can pass. Examples of flow cells and related fluidic detection systems and platforms that can be readily applied in the methods of the invention are described, for example, in Bentley et al., Nature 456: 53-59 (2008), in WO 04/018497; in US Pat. No. 7,057,026; in WO 91/06678; WO 07/123744; US Pat. Nos. 7,292,992; 7211414; 7315019; 7405281 and US Patent Application 2008/0108082.

[0035] As used herein, the term "amplicon", when referring to a nucleic acid, means a copy product of a nucleic acid, where the product has a nucleotide sequence that is identical or complementary to at least a portion of the nucleotide sequence of the nucleic acid. The amplicon can be generated by any of the various amplification methods that use a nucleic acid or its amplicon as a template, including, for example, polymerase extension, polymerase chain reaction (PCR), rolling ring amplification (RCA), extension by ligation or ligase chain reaction. An amplicon may be a nucleic acid molecule having a single copy of a particular nucleotide sequence (eg, a PCR product) or multiple copies of a nucleotide sequence (eg, an RCA concatemer product). The first amplicon of the target nucleic acid is usually a complementary copy. Subsequent amplicons are copies that are created after the production of the first amplicon from the target nucleic acid or from the first amplicon. The subsequent amplicon may have a sequence that is substantially complementary to the target nucleic acid or substantially identical to the target nucleic acid.

[0036] As used herein, the term "amplification site" means a site or array where one or more amplicons can be generated. The amplification site may also be configured to contain, retain or attach at least one amplicon that is generated at that site.

[0037] As used herein, the term "array" means a collection of sites that may differ from each other in their relative location. The difference between molecules that are in different positions of the array can differ from each other depending on the location of the sites in the array. A single array site may include one or more molecules of a particular type. Thus, for example, a site may include a single target nucleic acid molecule having a specific sequence, or the site may include multiple nucleic acid molecules having the same sequence (and/or complementary sequence). Array sites may have different features on the same substrate. Typical features include, but are not limited to, recesses in the substrate, spheres (or other particles) located in or on the substrate, protrusions from the substrate, folds in the substrate, or channels in the substrate. The sites of the array may be separate substrates, each of which contains different molecules. The various molecules attached to individual supports can be identified by the positions of the support on the surface to which they are attached, or by the positions of the support in a liquid or gel. Representative arrays in which individual substrates are located on the surface include, but are not limited to, arrays having spheres in wells.

[0038] As used herein, the term "capacity", when referring to a site and a nucleic acid molecule, means the maximum number of nucleic acid molecules that can occupy that site. Thus, for example, the term may refer to the total number of nucleic acid molecules that can occupy a site under specific conditions. Other parameters may also be used, including, for example, the total mass of nucleic acid molecules or the total number of copies of a particular nucleotide sequence that may occupy a site under particular conditions. Typically, the site capacity for the target nucleic acid will be substantially equivalent to the site capacity for the target nucleic acid amplicons.

[0039] As used herein, the term "capture agent" means a material, chemical, molecule, or portions thereof that are capable of attaching to, retaining on, or binding to a target molecule (eg, a target nucleic acid). Exemplary capture agents include, but are not limited to, a capture nucleic acid (also referred to herein as a capture oligonucleotide) that is complementary to at least a portion of the target nucleic acid; a member of a receptor-ligand binding pair (e.g., avidin, streptavidin), biotin, lectin, carbohydrate, nucleic acid-binding protein, epitope, antibody, etc.) capable of binding to (or binding to) the target nucleic acid a linker molecule attached to it), or a chemical reagent capable of forming a covalent bond with the target nucleic acid (or a linker molecule attached to it).

[0040] As used herein, the term "reporter moiety" can mean any identifiable label, marker, indexes, barcodes, or group that allows the composition, identity, and/or source of the analyte under investigation to be determined. In some embodiments of the invention, the reporter part may include an antibody that specifically binds to a protein. In some embodiments of the invention, the antibody may include a detectable label. In some embodiments of the invention, the reporter may include an antibody or affinity reagent labeled with a nucleic acid tag. The tagged nucleic acid can be detected, for example, by a proximal ligation assay (PLA) or a proximal extension assay (PEA) or a sequencing-based read (Shahi et al. Scientific Reports volume 7, Article number: 44447, 2017) or CITE-seq (Stoeckius et al. Nature Methods 14: 865-868, 2017).

[0041] As used herein, the term "clonal population" refers to a population of nucleic acids that is homogeneous with respect to a particular nucleotide sequence. The homogeneous sequence is usually at least 10 nucleotides in length, but may be even longer, for example, it may be at least 50, 100, 250, 500 or 1000 nucleotides in length. The clonal population can be derived from a single target nucleic acid or a template nucleic acid. Typically, all nucleic acids in a clonal population will have the same nucleotide sequence. It should be noted that a small number of mutations (eg, due to amplification artifacts) can occur in a clonal population without deviating from clonality.

[0042] As used herein, the term "unique molecular identifier" or "UMI" refers to a molecular label, either randomized, non-randomized, or semi-randomized, that can be attached to a nucleic acid. UMI, when incorporated into a nucleic acid, can be used to subsequently correct amplification bias by directly counting unique molecular identifiers (UMIs) that are sequenced after amplification.

[0043] As used herein, the term "exogenous" compound, eg, exogenous enzyme, refers to a compound that is not normally or inherently present in a particular composition. Thus, for example, if a particular composition includes a cell lysate, then an exogenous enzyme is an enzyme that is not normally or inherently present in the cell lysate.

[0044] As used herein, the term “providing” with respect to a composition, article, nucleic acid, or core means the receipt of a composition, article, nucleic acid, or core, the purchase of a composition, article, nucleic acid, or core, or the receipt of a compound, composition, article, or core by which in any other way.

[0045] The term "and/or" means one or all of the listed elements or a combination of any two or more of the listed elements.

[0046] The words "preferred" and "preferably" refer to embodiments of the invention that may provide certain advantages under certain circumstances. However, other embodiments of the invention may also be preferred under the same or different circumstances. In addition, when describing one or more preferred embodiments, it is not implied that other options cannot be used, and therefore other options cannot be excluded from the scope of the invention.

[0047] The terms "comprises" and their variants, if these terms are present in the description and in the claims, are not limiting.

[0048] It should be noted that wherever the embodiments of the invention described herein are used together with the word "comprise", "includes" or "including", etc., other similar embodiments described using the terms " consisting of" and/or "consisting essentially of".

[0049] Unless otherwise noted and indicated, the words "a", "an", "the" and "at least one" are synonymous and mean one or more than one.

[0050] In addition, in the present description, the indication of numerical ranges up to the boundary values includes all numbers included in this range (for example, the interval 1-5 includes the values 1; 1.5; 2; 2.75; 3; 3, 80; 4; 5, etc.).

[0051] For any method disclosed here that includes separate steps, these steps can be carried out in any possible order. And if necessary, two or more steps can be carried out simultaneously in any combination.

[0052] Throughout the description of the invention, the terms "one embodiment of the invention", "an embodiment of the invention", "certain embodiments of the invention" or "certain embodiments of the invention", and the like. mean that the specific features, configurations, compositions, or characteristics described in accordance with this embodiment are included in at least one embodiment of the disclosure. Thus, the use of such terms in various parts of the specification does not necessarily refer to the same embodiment of the disclosure. In addition, specific features, configurations, compositions, or characteristics may be combined in any suitable manner in one or more embodiments of the invention.

Brief description of the drawings

[0053] The following detailed description of exemplary embodiments of the present invention is provided for a better understanding of the invention when read with reference to the following drawings.

[0054] FIG. 1 is a general flow diagram of a general exemplary single cell combinatorial indexing method in accordance with the disclosure.

[0055] FIG. 2 is a general flow diagram of a general exemplary single cell combinatorial indexing method in accordance with the disclosure.

[0056] FIG. 3 is a general flow diagram of a general exemplary single cell combinatorial indexing method in accordance with the disclosure.

[0057] FIG. 4 shows that sci-RNA-seq3 can profile ~2 million cells from 61 mouse embryos at 5 developmental stages in a single experiment. (A) Working diagram of sci-RNA-seq3 and experimental scheme. (B) Comparison of experiment throughput with recent reports. (C) Scatter plot of mouse and human UMIs plotted for HEK293T and NIH/3T3 cells. (D) Histogram illustrating the number of cells profiled for each of 61 mouse embryos. (E) Box plot illustrating the number of genes and detected UMIs per cell. (F) Scatterplot of unique reads aligned to the Xist (chr X) and chrY genes per mouse embryo. (G) Pseudo-temporal trajectory of mouse embryonic RNA-seq pseudo-volume profiles. (H) Heat map of changes in E9.5-E13.5 marker gene expression from mouse embryonic RNA-seq pseudo-volume profiles.

[0058] FIG. 5 shows the results of trait assays and QC assays for sci-RNA-seq3. (A) Histogram illustrating the number of RT wells used for each of 61 mouse embryos. (B) Histogram illustrating the distribution of initial sequencing reads for each PCR well in sci-RNA-seq3. (C) Bar graph illustrating the number of UMI per cell for HEK293T and NIH/3T3 cells. (D) Box plot illustrating the ratio of reads mapped for expected HEK293T (human) and NIH/3T3 (mouse) cell types. (E) Box plot comparing UMIs per cell (downsampling to 20,000 original reads per cell) for sci-RNA-seq3 and sci-RNA-seq. (F) Correlation between measurements of gene expression levels in pooled HEK293T cell profiles for sci-RNA-seq3 and sci-RNA-seq.

[0059] FIG. 6 shows the results of additional trait assays and QC assays for sci-RNA-seq3. (A) Scatterplot illustrating the correlation between the number of OT wells used and the number of recovered cells per embryo. (B-D) Histogram for comparison of the number of original sequencing reads (B), detected genes (C) and UMI (D) per cell in sci-RNA-seq3 and other methods. (E) Bar graph illustrating the number of UMIs detected per embryo cell at all five developmental stages. (F) Histogram illustrating the number of male and female embryos profiled at each stage of development.

[0060] FIG. 7 shows RNA-seq pseudo-volume profiles of mouse embryos that have been easily separated into developmental stages. (A) t-SNE of pooled single cell transcriptomes obtained from each of the 61 mouse embryos identified five closely clustered groups perfectly aligned in developmental stages. (B) The pseudo-temporal trajectory of the mouse embryo RNA-seq pseudo-volume profiles is identical to the trajectory shown in FIG. 4G, but differs in that the pseudo time is shown in color. (C) Profiles of 61 embryos were sorted by pseudo-time. The three earliest and three latest (in pseudo-time) E10.5 embryos are shown in the photographs and differ in morphology.

[0061] In FIG. 8 illustrates the identification of major cell types involved in mouse organogenesis. (A) Visualization of t-SNE 2026641 mouse embryonic cells stained for cluster ID in Leuven clustering and annotated based on marker genes. The same t-SNE is shown in the graph below, where only cells at each developmental stage are illustrated. Primitive erythroid (transient) and definitive erythroid (reproducing) clusters are shown in the box to illustrate the dynamics of their proliferation. (B) Dot plot illustrating the expression of one selected marker gene per cell type. The spot size corresponds to the percentage of cells of each type, and its color corresponds to the average level of expression.

[0062] In FIG. 9 illustrates the identification of the major cell types involved in mouse organogenesis and the corresponding sets of marker genes specific to particular cell types. (A) Visualization of t-SNE as the cell transitions from E9.5 to E13.5, where developmental stages are shown in color. A similar t-SNE is shown in FIG. 8A, but in this case, each graph shows only the cells obtained at one time point. (B) Heat map illustrating relative gene expression across the major cell types identified. (C) Histogram illustrating the number of marker genes in each type of cell with more than twice the expression of the marker genes in the highest-expressing type 2 cells (FDR 5%).

[0063] FIG. 10 shows that cells originating from replicated embryos for the same time point show no overt periodic effects. (A-E) Visualization of t-SNE cells of mouse embryos at various stages of development: E9.5 (A), E10.5 (B), E1.5 (C), E12.5 (D), E13.5 (E ) stained for embryo ID at each stage.

[0064] FIG. 11 shows the dynamics of changes in the number of cells of specific types in the process of organogenesis in mice. (A) Histogram illustrating the number of cells profiled for each cell type by developmental stage. (B) Heat map illustrating the relative number of cells of each type (rows) in 61 mouse embryos (bars). The absolute number of cells of a particular type per embryo was calculated by multiplying the proportion that cells of a particular type would contribute to a given embryo by the total calculated number of cells at that stage of development. For data presentation, these scores are normalized in each row to the maximum estimated number of cells of a given type for a total of 61 embryos. Embryos were sorted from left to right by pseudo-development time. (C) Line graph illustrating the relative change in cell number for primitive erythroid and definitive erythroid lineages calculated as in panel B. The dashed lines show the relative expression of marker genes for primitive erythroids (Hbb-bhl) and definitive erythroid lineages (Hbb-bs). Experimental data for individual embryos were sorted by pseudo-development time and smoothed by the Less method.

[0065] FIG. 12 illustrates Leuven clustering and visualization of t-SNE subclusters of each of the 38 major cell types. Since the heterogeneity of cell types was quite evident for the set of 38 clusters shown in FIG. 8A, we developed an iterative Leuven iterative clustering strategy for each major cell type in order to identify subclusters. After removing subclusters dominated by one or two embryos, and after merging very similar subclusters, a total of 655 subclusters (also called "subtypes" to distinguish them from the 38 main cell types identified by initial clustering) were obtained.

[0066] FIG. 13 illustrates Leuven clustering and visualization of t-SNE subclusters of each of the 38 major cell types at all developmental stages. This figure is identical to Fig. 12, except that the cells were stained for developmental stage and not for subcluster ID.

[0067] FIG. 14 illustrates the sensitivity of cell type detection as a function of cell coverage. (A) t-SNE imaging of all cells (left plot, n=2026641) and lower sampled subset (right plot, n=50000) stained for Leuven cluster IDs as shown in FIG. 8A. (B) t-SNE imaging of all endothelial cells, (left panel, n=35878) and lower sampled subset cells (right panel, n=1173), stained for Leuven cluster ID and computed from 35878 endothelial cells. (C) Visualization of t-SNE of 1173 endothelial cells stained for the Leuven Cluster ID and computed from 1173 endothelial cells.

[0068] FIG. 15 illustrates 655 cell subtypes derived from a variety of embryos and defined by a series of markers. (A) Histogram illustrating the distribution of subclusters by cell number (median 1869; range 51-65894). (B) Histogram illustrating the distribution of subclusters by the number of embryos from which these cells originate (>5 cells qualified as a favorable factor). (C) Histogram illustrating the distribution of subclusters in relation to cells derived from the embryo considered as the most favorable factor. (D) Histogram illustrating the distribution of subclusters by number of marker genes (with an expression level that is at least 1.5 times the level of expression compared to the level of expression in cells of other subtypes with the highest level of expression in the same main cluster; 5% FDR).

[0069] FIG. 16 shows the dynamics of changes in the number of cells of specific types during organogenesis in mice. (A) Heat map illustrating the relative gene expression of the 655 identified subclusters. (B) Heat map illustrating the relative number of cells of each subtype (rows) in 61 mouse embryos (bars). The absolute number of cells of a particular subtype per embryo was calculated as shown in FIG. 11V. (C) Visualization of t-SNE of all 61 mouse embryos based only on the proportion of 655 cell subtypes in each embryo.

[0070] FIG. 17 illustrates the identification and characterization of epithelial cell and apical ectodermal limb ridge (AER) subtypes. (A) Visualization of t-SNE and annotation of epithelial cells of certain subtypes based on markers. (B) Visualization of t-SNE of all epithelial cells stained for Fgf8 expression level. (C) Images of Fgf8 in situ hybridization lanes in E10.5 (left) and E11.5 (right) embryos. (D) Visualization of t-SNE of all epithelial cells stained for Fndc3a expression levels. (E) Images of Fndc3a in situ hybridization lanes in an E10.5 embryo. Arrow: gene expression site. (F) Bar graph illustrating the proportion of AER cells per embryo at various stages of development. (G) Pseudo-temporal trajectory of AER single-cell transcriptomes stained for developmental stages. (H) Line graph illustrating the relative expression of AER marker genes during the entire developmental pseudo-time.

[0071] In FIG. 18 illustrates the identification of mouse epithelial subtypes. Dot plot illustrating the expression of one selected marker gene per epithelial subtype. The spot size corresponds to the percentage of cells of a certain type, and its color corresponds to the average level of expression.

[0072] FIG. 19 illustrates the dynamics of gene expression in apical ectodermal limb ridge (AER) cells over pseudotime. (A) Heat map illustrating smoothed pseudo-time differential gene expression (FDR 1%) in AER cells plotted by negative binomial regression and scaled by percentage of maximum gene expression. Different genes are indicated in each row, and these genes are distributed into subsets that have an activated (top), repressed (middle) or transient (bottom) dynamics between E9.5 and E13.5. (B-C) Plots illustrating the q-value converted to log10 and pooled scores for enriched reactome members calculated with enrichR (B) and transcription factors (C) for genes that are significantly downregulated by AER. The best members of the enrichment pathway (Reactome2016) with a significant reduction in the number of genes include the passage of the cell cycle (mitotic cell cycle, qva1=0.0002) and glucose metabolism (carbohydrate metabolism, qva1=0.0002). The most enriched TFs with gene reduction targets include pluripotency factors such as Is11 (qva1 <10-5), Pou5f1 (qva1=0.002) and Nanog (qva1=0.003).

[0073] FIG. 20 illustrates the characterization of cell trajectories during limb mesenchymal differentiation. (A) 3D UMAP imaging of mesenchymal limb cells stained for developmental stages (bidirectional images are shown on the left and right). (B) Scatterplot illustrating normalized Pitxl and Tbx5 expression in limb mesenchymal cells. Only cells in which Pitxl and/or Tbx5 are found are shown. (C) Volkano plot illustrating the differentially expressed genes (FDR 5%, shown in red) between the fore and hind limbs. Genes with the highest level of differential expression were labeled. X-axis: log2-transformed fold change between fore and hind limbs for each gene. Y-axis: - log10-transformed qval from the differentiation test. (D) Same imaging as panel A, but staining for normalized gene expression for proximal markers/chondrocytes (Sox6, Sox9) and markers for distal organs (Hoxdl3, Tfap2b), anterior (Pax9, Alx4), or posterior (Hand2, Shh) limbs. (F) Images of in situ hybridization lanes for Hoxdl3 in E10.5-E13.5 embryos. (G) Same visualization as in panels A and D, but stained for normalized Cpa2 gene expression. Their expression pattern along this trajectory allowed the authors to predict that Cpa2 is a distal marker of limb mesenchymal development, such as Hoxdl3. (H) Images of in situ hybridization lanes for Cpa2 in E10.5-E11.5 embryos. (I) Pooled systematic trajectory results for AER and limb mesenchyme.

[0074] FIG. 21 illustrates the characterization of cell trajectories during limb mesenchymal development. (A) Heat map illustrating the genes with the highest levels of differential expression between different stages of limb mesenchymal cell development. (B) Histogram illustrating the log10-transformed adjusted p-value for enriched transcription factor genes with increased activation during limb mesenchymal development. (C) Visualization of t-SNE mesenchymal limb cells stained for forelimb (Tbx5+) and hindlimb (Pitx1+). Cells lacking or expressing both Tbx5 and Pitxl are not shown.

[0075] FIG. 22 shows the expression of markers spatially restricted in limbs. Each panel illustrates a different marker gene. Color shows UMI numbers that have been scaled to library size, log-transformed, and then mapped to Z-scores for comparison between genes. Cells with no expression of this marker were excluded to prevent overshooting of the points on the graph. (A) Pitxl hind limb marker and Tbx5 forelimb marker. (B) First row: proximal limb markers Sox6 (which also mark chondrocytes) and Sox9. Second row: distal limb markers Hoxdl3 and Tfap2b. Third row: forelimb markers 68 Pax9 and Alx4. Fourth row: hindlimb markers Shh and Hand2.

[0076] FIG. 23 shows modules of spatially restricted genes in limbs. A total of 1191 genes were clustered using hierarchical clustering. The dendrogram was divided into 8 modules using the tree boundary function in R, and the cumulative gene expression in each module was evaluated. The color shows the cumulative UMI values for each module, which were scaled by library size, log-transformed, and then mapped by Z-score for comparison between modules. Cells with no expression of this module were excluded to prevent exceeding the points on the graph.

[0077] FIG. 24 illustrates the characterization of eight major developmental trajectories observed during organogenesis in mice. (A) 3D UMAP visualization of the authors' total dataset; top: two-way view; Bottom: Enlarged view of mesenchymal (left) and neural tube/notochord (right) trajectories, stained by developmental stage. (B) Heatmap illustrating the proportion of cells from each of the 38 major cell types assigned to each of the 8 major trajectories. The columns represent the eight major lines of differentiation, shown in color in the upper histogram (see key in panel A). (C) 3D UMAP visualization of epithelial subtrajectories stained for developmental stages.

[0078] FIG. 25 illustrates the characterization of eight major developmental trajectories observed during organogenesis in mice. (A) This figure is similar to Fig. 24A, except that the color corresponds to the 38 main cell clusters. (B-C) Plot area showing the estimated proportion (B) and the estimated absolute number of cells (C) per embryo obtained for each of the eight major cell trajectories from E9.5 to E13.5.

[0079] FIG. 26 illustrates a UMAP visualization of eight major cell trajectories stained for the major cell cluster ID.

[0080] FIG. 27 illustrates UMAP imaging for eight major cell trajectories, stained by developmental stage.

[0081] FIG. 28 illustrates the visualization of UMAP epithelial cell subtypes. The stain corresponds to the 29 epithelial subtypes shown in FIG. 17A.

[0082] FIG. 29 shows the resolution of cell trajectories in myogenesis. The edges on the main graphs that define the trajectories reported to Monocle 3 are shown as blue patch bars. (A) Cells putatively involved in myogenesis were isolated from the mesenchymal cell trajectory in silico and then used to create a myocyte subtrajectory (Methods). (B) Cells in a myocyte subtrajectory stained for developmental stages. (C) Cells in the myocyte trajectory stained for the expression of selected transcriptional regulators during myogenesis. Cells without detectable expression for this gene were excluded from the graph. (D) Cells classified by developmental stages according to markers are shown in panel C (Methods).

[0083] FIG. 30 illustrates joint profiling of the total and newly synthesized transcriptome for sci cell fate. (A) Working diagram of sci cells with key steps outlined in the description. (B) Schematic of the experiment. A549 cells were treated with dexamethasone in a time dependent manner. Cells under all treatment conditions were labeled with S4U two hours prior to harvest to assess sci cell fate. (C) Violin plot showing the ratio of S4U-labeled reads per cell over six treatments. (D) Violin plot showing the ratio of S4U-labeled reads in exon and intron reads. For all rectangular charts: thick horizontal lines; medians; the top and bottom edges of the rectangle; first and third quartiles, respectively; whiskers, 1.5 times the interquartile interval; circles, outliers. (E) Visualization of UMAP in A549 cells using whole transcriptome (left), newly synthesized transcriptome (middle) and both (right). (F) Visualization similar to (E), but staining for cluster ID detected using the whole transcriptome. (G) Visualization of UMAP in A549 cells by common information staining for normalized G2/M marker gene expression at the RNA level (left) and at the newly synthesized RNA level (right). The UMIs for these genes are scaled by library size, log-transformed, pooled, and then mapped to Z-scores.

[0084] FIG. 31 shows the results of additional trait assays and QC assays for sci cell fate. (A) Scatterplot of murine (NIH/3T3) and human (HEK293T) UMIs per cell under sci cell fate assessment conditions. (B-D) Bar graph showing ratio of S4U-labeled reads, UMI number, and purity (proportion of reads mapped to expected species) per cell for HEK293T cells (n=932 cells) and NIH/3T3 cells (n=438 cells) ). For all rectangular charts: thick horizontal lines; medians; the top and bottom edges of the rectangle; first and third quartiles, respectively; whiskers, 1.5 times the interquartile interval; circles, outliers. (E-F) Correlation (Spearman correlation) between scores of gene expression measurements in pooled HEK293T (E) and NIH/3T3 (F) cell profiles by sci-fate (y-axis) versus sci-RNA-seq cells (x-axis) .

[0085] FIG. 32 shows the sci fate of A549 cells treated with dexamethasone. (A, B) Violin plot showing the number of UMIs (A) and genes (B) per cell under six treatment conditions. For all rectangular charts: thick horizontal lines; medians; the top and bottom edges of the rectangle; first and third quartiles, respectively; whiskers, 1.5 times the interquartile interval; circles, outliers. (C) Correlation plot showing the Pearson correlation coefficient between different processing conditions for the pooled whole transcriptome (upper right) and the newly synthesized transcriptome (lower left). (D) Visualization of UMAP in A549 cells using the newly synthesized transcriptome, staining for the cluster ID identified using the newly synthesized transcriptome. (E) Heat map showing the proportion of cells from each cluster identified across the entire transcriptome that are included in each cell cluster of the newly synthesized transcriptome. (F-G) Visualization of UMAP in A549 cells using total and newly synthesized transcriptome stained for normalized expression of S-phase marker genes for total RNA expression (F) and newly synthesized RNA (G). The UMIs for these genes are scaled by library size, log-transformed, pooled, and then mapped to Z-scores.

[0086] FIG. 33 illustrates the characterization of TF modules that initiate the transition of cells from one state to another. (A) Identified relationship (blue) between transcription factors (orange) and regulated genes (grey). TF modules associated with cell cycle progression or GR response are labeled. (B) UMAP imaging of A549 cells sorted into cell cycle TF modules with staining of newly synthesized mRNA for S-phase and G2/M-phase markers (top), three cell cycle phases (lower left), and nine cell cycle stages by analysis on uncontrolled clustering (bottom right). (C) Visualization of UMAP A549 cells sorted by TF GR response modules staining for DEX treatment time (left), CEBPB and FOXO1 activity (middle), and cluster ID in the uncontrolled clustering assay (right). To calculate TF activity, the number of newly synthesized UMIs for these genes is scaled by library size, log-transformed, pooled, and then Z-score mapped. (D) Table showing the observed relationship (in black) of cell state to the combinatorial state of cell cycle modules (x-axis) and GR response modules (y-axis). The red number means the expected relationship, assuming an independent set. (E) Heat map illustrating the proportion of cell states determined from the combinatorial states of the TF modules in each of the major clusters identified in the pooled whole and newly synthesized transcriptome clustering assay.

[0087] FIG. 34 illustrates TF modules initiating cell transition from one state to another for DEX-treated A549 cells. (A) Identified target genes (grey) from CEBPB (orange). Only relationships with adjustable correlation coefficient based on LASSO >0.6 are shown. (B) UMAP imaging of A549 cells on the whole and newly synthesized transcriptome, staining for CEBPB expression (left) and activity (right). (C) Similar to (B) but stained for YOD1 expression (left) and YOD1 activity (right). (D) Similar to (B), but stained for GTF2IRD1 expression (left) and for GTF2IRD1 activity (right). (E) Similar to (B), but stained for E2F1 expression (left), for E2F1 activity (middle), and for overall whole-transcriptome expression for E2F1-linked genes (right). (F) Heat map showing the absolute value of the Pearson correlation coefficient between TF modules. The 29 TF modules were divided into five groups using a hierarchical clustering analysis.

[0088] FIG. 35 shows cellular states characterized by combinatorial states of TF functional modules. (A) Scheme illustrating the strategy for characterizing cell states by combinatorial states of TF functional modules. (B) Umap imaging for all cells using both whole and newly synthesized transcriptome stained for the main cluster identifier identified by the UMAP region density peak clustering algorithm.

[008 9] FIG. 36 illustrates the characterization of cell transition trajectories from one state to another for >6000 individual cells. (A) Schematic illustrating memory correction and cell fusion analysis to construct a cell transition trajectory from one state to another, discussed in detail in the description and Method. (B) 3D plot of cells stained by DEX treatment time (also as z-coordinates). The x and y coordinates correspond to the UMAP region of the whole and newly synthesized transcriptome in FIG. 30E (left). Merged parent and child cells are shown with gray lines. (C) Similar to (B), except that the x and y coordinates correspond to the UMAP region in single cell monocellular transcriptome dynamics at six time points. (D) Line graphs illustrating the dynamics of cell state in terms of different GR responses (top) and cell cycle phase (bottom) in clusters for each cell trajectory (left) or in all cells (right) regardless of cell fusion assay. (E) Network of cell transition from one state to another. The nodes are the 27 cell states described in FIG. 33D, and links are identified pathways for cells to move from one state to another. Links with a low transition probability (<0.1) are filtered out. Squares with dotted lines show examples of states with reversible transition dynamics. (F) Correlation plot illustrating the correlation of cell state proportion between treatment conditions. Positive correlations are shown in blue and negative ones in red. The shape of the ellipse correlates with the correlation coefficients (on the ellipse). (G) Scatterplot illustrating the correlation of cell state fractions between observed 10-hour DEX treatment groups and predicted cell state fractions. The prediction is based on the probabilities of cell transition from one state to another and on an estimate of the proportion of cell state in the group without DEX treatment. The blue line corresponds to the linear regression line. (H) Scatterplot illustrating the correlation of cell transition probability from one state to another calculated from complete data (0-10 hours) or partial data (0-6 hours) along with a linear regression line.

[0090] FIG. 37 shows an assessment of the level of detection of new RNA and the level of degradation of RNA. (A) Scatterplot illustrating the correlation between x-axis: differences in normalized whole transcriptome between cells without DEX and cells 2 hours after DEX treatment, and y-axis: differences in normalized newly synthesized transcriptome between cells without DEX and cells 2 hours after DEX treatment . The blue line indicates the linear regression line. The complete transcriptome and the newly synthesized transcriptome at each time point are normalized to the library size of the entire transcriptome at that time point. (B) Correlation plot illustrating the correlation of the gene degradation score between treatment conditions. Positive correlations are shown in blue and negative ones in red. The shape of the ellipse correlates with the correlation coefficients (on the ellipse).

[0091] FIG. 38 shows a cell transition network from one state to another to predict the state of the cells. (A) Correlation plot illustrating the correlation between observed cell states at each treatment time and predicted cell state by cell transition probability from one state to another and by proportion of cell states in the non-DEX treatment group. The blue line shows the linear regression line. (B) Scatterplot illustrating the correlation of the proportion of cell states between observed 10-hour DEX treatment groups and predicted values. The predicted values are based on the probabilities of transition of cells from one state to another, estimated from partial data (0-6 hours), and on the proportion of cell state in the group without DEX treatment. The blue line shows the linear regression line.

[0092] FIG. 39 shows that the probabilities of cell transition from one state to another are regulated by the transition stability pattern to the neighboring state. (A) Correlation plot illustrating the correlation between transition distance (Pearson distance) and cell transition probability from one state to another, along with the red Lesse smoothing line labeled ggplot2. (B) 3D graph illustrating the pattern of cell state instability. The x-axis represents the states of the GR responses (from "none" to "low" to "high"). The y-axis represents the cell cycle phases from G0/G1 to G2/M. The z-axis represents the cell state instability pattern, defined by the probability of each cell state jumping to other states after 2 hours. (C) Scatterplot illustrating the relationship between cell state instability and cell fraction change before and after 10-hour DEX treatment, along with the red Loess smoothing line labeled ggplot2. (D) Scatterplot illustrating the correlation between state instability and state transition entropy along with a linear regression line (in blue). (E) Box plot illustrating R-squared cross-validation analysis for predicting state transition probability from transition distance alone or from a combination of transition distance and state instability pattern using a densely connected neural network.

[0093] FIG. 40 is a scan of the target exon in the LMO2 gene. The target exon is listed on the Target Exons track. It should be noted that the 12 RT primers passed through the filters cover the exon, if possible, at a sufficient level. The bottom two lanes, "Primers_plus/minus" and "Captured_plus/minus", illustrate reads mapped to RT primers and captured transcripts, respectively. The difference in the number of reads indicates the number of unwanted events triggered.

[0094] FIG. 41 is a table for the comparative ranking of genes in the ENCODE total nuclear RNA sequencing database and genes in the in situ multiplex RT capture library. Target genes are shown in bold. Of the 12 most common genes in the target library, 8 target genes were selected. In the last two columns - "RANK" means the ranking in the database of inventors, and "ENCODE" means the ranking in the ENCODE database. Excess non-target RNAs include the mitochondrial ribosomal RNAs RNR2 and RNR1, and frequently occurring nuclear RNAs include IncRNA MALAT1. It should be noted that the LMO2 gene ranges from the 4627th most detected gene in the ENCODE database (out of 26281 genes) to the 3rd most detected gene in the authors' library.

[0095] FIG. 42 shows the enrichment of the captured transcript compared to total RNA. Scatterplot for reads in the target library and for reads in ENCODE nuclear RNA. Genes targeted by OTs are shown in red; however, it should be noted that they are usually off the diagonal, indicating that they were enriched compared to the expected level of expression in the reference database. Excess nuclear IncRNA MALAT1 and XIST are shown in blue: and they are above the diagonal, indicating successful enrichment compared to these RNAs. The target eRNA is not enriched (orange). The mean enrichment was 45.3 times the expected level for the 9 target genes in this experiment. Overall, target gene reads account for 31% of the total reads mapped by genes in this database.

[0096] The schematic drawings are not necessarily to scale. Like numbers used in the figures refer to like components, steps, and the like. However, it should be noted that the use of a number to refer to a component in this drawing is not to be considered as limiting the component in another drawing designated by the same number. In addition, the use of different numbers to designate components does not mean that different numbered components cannot be identical or similar to other numbered components.

Detailed description of illustrative embodiments of the invention

[0097] In one embodiment of the invention, the method described here can be used to create single cell sequencing (sci) and combinatorial indexing libraries that include multiple single cell transcriptomes. For example, this method can be used to obtain information about the sequences of whole cell transcriptomes, transcriptomes of newly synthesized RNA, or a combination of both. In another embodiment of the invention, the method described here can be used to obtain sci-sequencing libraries that include information about the sequence of a subset of RNA nucleic acids. Thus, for example, if a non-coding regulatory region is the target of a permutation, then a coding region that is cis to the regulatory region can be tested for altered expression. In another example, cell mapping experiments can be performed using mRNA-limited readouts that are highly informative.

[0098] The method may include one or more steps of obtaining isolated nuclei or cells, distributing subpopulations of the isolated nuclei or cells into compartments, treating the isolated nuclei or cells to include nucleic acid fragments, and adding a compartment-specific index to the nucleic acid fragments. This method may include, but not necessarily, the processing of cells under predetermined conditions and/or labeling of newly synthesized RNA in cells. This method may be directed to obtaining information that includes information about the cellular transcriptome or subpopulation of RNA nucleic acids. These steps can be performed in almost any order and can be combined in a variety of ways. Optionally, nuclei can be isolated from cells after treating the cells under predetermined conditions and labeling the newly synthesized RNA.

Obtaining isolated nuclei or cells

[0099] The method described here may include obtaining cells or nuclei isolated from a plurality of cells (Fig. 1, block 10; Fig. 2, block 22). Cells can be derived from any organism(s) and from any cell type or tissue of the organism(s). In one embodiment of the invention, the cells may be embryonic cells, such as cells derived from an embryo. In one embodiment of the invention, cells or nuclei can be isolated from cancerous or diseased tissue. The method may also include cell dissociation and/or isolation of nuclei. The number of nuclei or cells may be at least two. The upper limit depends on the practical limitations of the equipment (eg, multiwell plates, number of indexes) used in other steps of the method described here. The number of nuclei or cells that can be used is not particularly limited and may be in the billions. So, for example, in one embodiment of the invention, the number of nuclei or cells can be no more than 100000000, no more than 10000000, no more than 1000000000, no more than 100000000, no more than 10000000, no more than 1000000, no more than 100000, no more than 10000, no more 1000, not more than 500, or not more than 50. It will be appreciated by those skilled in the art that in some embodiments of the invention, the nucleic acid molecules in each nucleus represent the entire transcriptome of that nucleus, such as the entire transcriptome, a newly synthesized transcriptome, or both.

[00100] In embodiments where isolated nuclei are used, nuclei can be obtained by extraction and fixation. Optionally and preferably, the method for obtaining isolated nuclei does not include enzymatic treatment. In those embodiments where a newly synthesized transcriptome is obtained, nuclei are not isolated until the cell has been treated under conditions suitable for labeling the newly synthesized transcriptome.

[00101] In one embodiment of the invention, nuclei are isolated from single cells that are adherent or present in suspension. Methods for isolating nuclei from individual cells are known to the person skilled in the art. Nuclei are usually isolated from cells present in the tissue. The method for producing isolated nuclei typically involves obtaining tissue, isolating nuclei from the resulting tissue, and then fixing the nuclei. In one embodiment of the invention, all steps are carried out on ice.

[00102] Tissue preparation involves flash freezing the tissue in liquid nitrogen and then reducing the size of the tissue by cutting it into pieces with a diameter of 1 mm or less. Tissue can be reduced in size by shredding or by physical manipulation. Grinding can be done by using a razor blade to cut the tissue into small pieces. The application of physical impact can be carried out by breaking the fabric with a hammer or the like, and the resulting composition of the crushed fabric is called a powder.

[00103] Isolation of nuclei can be accomplished by incubating pieces of tissue or powder in cell lysis buffer for at least 1-20 minutes, such as 5, 10, or 15 minutes. Suitable buffers are buffers that promote cell lysis but maintain the integrity of the nuclei. An example cell lysis buffer includes 10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl ₂ , 0.1% IGEPAL CA-630, 1% SUPERase in RNase inhibitor (20 U/µl, Ambion) and 1% BSA (20 mg/ml, NEB). Standard methods for isolating nuclei often use one or more exogenous compounds, such as exogenous enzymes, to facilitate isolation. Examples of useful enzymes that may be present in cell lysis buffer include, but are not limited to, protease inhibitors, DNase, lysozyme, proteinase K, surfactants, lysostaphin, zymolase, cellulase, protease or glycanase, and the like. (Islam et al., Micromachines (Basel), 2017, 8(3):83; www.sigmaaldrich.com/life-science/biochemicals/biochemical-products.html?TablePage=14573107). In one embodiment of the invention, one or more exogenous enzymes are not present in the cell lysis buffer used in the method described here. For example, an exogenous enzyme is (i) not added to cells prior to mixing cells and lysis buffer, (ii) not present in cell lysis buffer prior to mixing with cells, (iii) not added to the mixture of cells and cell lysis buffer , or combinations thereof. One of ordinary skill in the art will appreciate that these component levels can be altered somewhat without compromising the value of cell lysis buffer for nuclei isolation. The extracted nuclei are then purified by performing one or more rounds of washing with nuclear buffer. An example nuclear buffer includes 10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl ₂ , 1% SUPERase in RNase inhibitor (20 U/μl, Ambion) and 1% BSA (20 mg/ml, NEB ). Like cell lysis buffer, exogenous enzymes may also be absent from the nuclear buffer used in the method of the invention. It will be apparent to those skilled in the art that these component levels can be altered somewhat without degrading the value of the nuclei extraction buffer. It will also be apparent to those skilled in the art that BSA and/or surfactants can be used in buffers used to isolate nuclei.

[00104] The isolated nuclei are fixed by exposure to a cross-linking agent. A suitable example of a cross-linking agent includes, but is not limited to, paraformaldehyde. Paraformaldehyde may be present at a concentration of 1% to 8%, for example 4%. Treatment of nuclei with paraformaldehyde may include the addition of paraformaldehyde to the suspension of nuclei and incubation at 0°C. Optionally, but preferably, a wash in nuclear buffer is performed after fixation.

[00105] The isolated fixed nuclei can be used directly or aliquoted and then immediately frozen in liquid nitrogen for later use. In preparation for use after freezing, thawed kernels can be made permeable, for example by using 0.2% Triton-100 for 3 minutes on ice, and then rapidly sonicated to reduce kernel accumulation.

[00106] Standard techniques for extracting nuclei from tissues typically involve incubating the tissue with a tissue-specific enzyme (eg, trypsin) at high temperature (eg, 37°C) for a period of 30 minutes to several hours, and then subjecting the cells to lysis with a buffer to cell lysis to extract nuclei. The nuclear isolation method described here has several advantages: (1) no artificial enzymes are introduced and all steps are performed on ice. This leads to a possible reduction in the perturbation of cell states (eg transcriptome states). (2) The new method has been tested on most tissue types, including brain, lung, kidney, spleen, heart, cerebellum tissues, and abnormal tissue samples such as tumor tissue. Compared to traditional methods for extracting nuclei from tissues, which use different enzymes for different types of tissues, the new method can potentially reduce bias when comparing the states of cells from different tissues. (3) The new method also reduces cost and improves efficiency by eliminating the enzyme treatment step. (4) Compared with other methods of extraction of nuclei (for example, the Downs method of tissue grinding), the new method is more reliable for various types of tissues (for example, the Downs method requires optimization of Downs cycles for various tissues) and can process large sample fragments with high throughput (e.g. the Downs method is limited by the size of the grinder).

[00107] The isolated nuclei may optionally not contain a nucleosome, or may be placed under conditions that deplete nuclei made up of nucleosomes, resulting in nuclei depleted of nucleosomes.

Distribution of subpopulations

[00108] The method described here includes distributing subpopulations of isolated nuclei or cells across multiple compartments (FIG. 1, block 11; FIG. 2, block 23; FIG. 3, block 32). The method may include a plurality of distribution steps where a population of isolated nuclei or cells (also referred to herein as a pool) is divided into subpopulations. Typically, subpopulations of isolated nuclei or cells, such as subpopulations present in multiple compartments, are indexed with compartment-specific indices and then pooled. Accordingly, the method typically includes at least one "split and pool" step consisting of collecting the isolated nuclei or cells, distributing them, and adding a compartment-specific index, where the number of "split and pool" steps may depend on the number of different indices. , which are added to nucleic acid fragments. Each initial subpopulation of nuclei or cells before indexing may be unique among other subpopulations. For example, each first subpopulation may be taken from a unique sample or processed under unique conditions. After indexing, the subpopulations can be pooled, subpopulated, indexed, and pooled again, as needed, until enough indexes have been added to the nucleic acid fragments. This process allows you to assign a unique index or combination of indexes to each individual cell or nucleus. After indexing is complete, for example after adding one, two, three or more indexes, the isolated nuclei or cells can be lysed. In some embodiments of the invention, the addition of the index and lysis can occur simultaneously.

[00109] The number of nuclei or cells present in a subpopulation, and therefore in each compartment, may be at least 1. In one embodiment of the invention, the number of nuclei or cells present in a subpopulation is no more than 100,000,000, no more less than 10000000, not more than 1000000, not more than 100000, not more than 10000, not more than 4000, not more than 3000, not more than 2000 or not more than 1000, not more than 500 or not more than 50. In one embodiment of the invention, the number of nuclei or cells present in a subpopulation may be from 1 to 1000, from 1000 to 10000, from 10000 to 100000, from 100000 to 1000000, from 1000000 to 10000000 or from 10000000 to 100000000. In one embodiment of the invention, the numbers of nuclei or cells present in each subpopulation are approximately the same. The number of nuclei present in a subpopulation, and therefore in each compartment, was partly based on the desire of the authors to reduce index overlap, which is the presence of two nuclei or cells having the same combination of indices ending in the same compartment at this stage of the method. Methods for distributing nuclei or cells into subpopulations are known to the person skilled in the art and are routine. While fluorescence-activated cell sorting (FACS) cytometry can be used, in some embodiments, simple dilution is preferred. In one embodiment of the invention, FACS cytometry is not used. Nuclei having different ploidies can be gated and enriched with staining, such as DAPI (4',6-diamidino-2-phenylindole) staining. Staining can also be used to separate single cells from doublets during sorting.

[00110] The number of compartments in the distribution steps (and subsequent index addition) may depend on the format used. For example, the number of compartments can be from 2 to 96 compartments (using a 96-well plate), from 2 to 384 compartments (using a 384-well plate), or from 2 to 1536 compartments (using a 1536-well plate). In one embodiment of the invention, multiple tablets may be used. In one embodiment of the invention, each compartment may be a drop. If the type of compartment used is a drop that contains two or more nuclei or cells, then any number of drops can be used, such as at least 10,000, at least 100,000, at least 1,000,000, or at least 1,000,000 drops. Subpopulations of isolated nuclei or cells are usually indexed into compartments prior to pooling.

[00111] In some embodiments, the compartment is a drop or well. A transcriptome, a newly synthesized transcriptome, or subpopulations thereof in a cell or nucleus can be labeled with a unique index or combination of indices in a droplet or well. Indexed libraries isolated from separate parts of drops or wells can be pooled for further processing and sequencing. Examples of such methods include, but are not limited to, single cell analysis systems from 10X Genomics (Pleasanton, CA), Biorad (Hercules, CA), and CellSee (Ann Arbor, MI).

Processing under predefined conditions

[00112] In an optional embodiment of the invention, each subpopulation of cells is exposed to an agent or perturbation (Fig. 1, block 12). Such an agent can be essentially any agent that causes a change in a cell. For example, an agent can change the transcriptome of a cell, change the structure of the cell's chromatin, change the activity of a protein in the cell, change the cell's DNA, change the methylation state, change the editing of the cell's DNA, or cause other changes. Examples of agents include, but are not limited to, a compound such as a protein (including an antibody), a non-ribosomal protein, a polyketide, an organic molecule (including an organic molecule of 900 Daltons or less), an inorganic molecule, an RNA or RNAi molecule, a carbohydrate, glycoprotein, nucleic acid, or a combination thereof. In one embodiment of the invention, the agent causes a genetic perturbation, for example, a DNA editing protein such as CRISPR or Talen. In one embodiment of the invention, the agent is a therapeutic drug. In one embodiment, the cell may be a wild-type cell, and in another embodiment, the cell may be genetically modified to include a genetic perturbation, such as a gene knockin or gene knockout (Szlachta et al., Nat Commun. , 2018, 9:4275). Subpopulations of cells can be exposed to the same agent, but different parameters can change in different compartments, allowing multiple parameters to be tested in a single experiment. For example, different doses, different treatment times and different cell types can be tested in one multiwell plate. In one embodiment of the invention, the cells can express a protein with a known activity, and the effect of the agent on the activity is evaluated under various conditions. The use of index sequences to label nucleic acid fragments then allows the identification of nucleic acids derived from specific subpopulations of nuclei or cells, for example, from one well of a multiwell plate.

[00113] Labeling of nucleic acids

[00114] In an optional embodiment of the invention, nucleic acids such as RNA, cDNA or DNA produced by the cell were labeled (FIG. 1, block 13). Modern methods of genomic technologies for individual cells provide an instant image of the state of the cell, but do not provide information about the dynamics of the cell transition. The present inventors have found that labeling newly synthesized RNA allows capturing both the entire transcriptome and the newly synthesized transcriptome at the individual cell level by a split-and-combine indexing method, a combinatorial indexing method, or any single cell indexing method. The entire transcriptome and newly synthesized RNA receive the same unique index or combination of indices, which makes it possible to identify existing (for example, pre-existing) and newly synthesized nucleic acids in the same cell. This allows characterization of the dynamics of cell transition from one state to another, regulated by internal factors (eg, the internal cell cycle program) and external factors (eg, the cell's response to an external stimulus, such as a therapeutic drug). In addition, in some embodiments, the capture of both the entire transcriptome and the newly synthesized transcriptome at the single cell level is provided, along with information about the decomposition of the transcriptome compared to its previous state (past state memory). Each cell's memory of past states can be adjusted for the degree of mRNA degradation (memory correction), so that each cell can be characterized by transcriptome dynamics between two or more time points.

[00115] There are various methods for labeling a newly synthesized nucleic acid so that it can be distinguished from an already existing nucleic acid, and basically any method can be used. Typically, the label is incorporated into nucleic acids as they are synthesized. One type of method involves the introduction of a nucleoside analog that adds an identifiable mutation. For example, the addition of the nucleoside analog 4-thiouridine (S4U) to an RNA molecule leads to a point mutation during the reverse transcription step, and thus, to the formation of a mutated cDNA of the first strand, which has a thymine-to-cytosine substitution (Sun and Chen, 2018, Metabolic Labeling of Newly Synthesized RNA with 4sU to in Parallel Assess RNA Transcription and Decay, In: Lamande S. (eds) mRNA Decay, Methods in Molecular Biology, vol. 1720, Humana Press, New York, NY). This point mutation can be identified at the sequencing and analysis steps by comparing the sequence with a reference. Another type of method involves the introduction of a hapten-labeled nucleotide, which can be used to purify RNA containing the hapten. Examples are biotinylated nucleotides (Luo et al., 2011, Nucl. Acids Res., 39 (19): 8559-8571) and digoxigenin modified nucleotides (supplied from Jena Bioscience GmbH). A third type of method involves introducing a nucleotide that can be modified by a chemical reaction, such as a nucleotide functionalized by pressing a button on a machine, and adding a hapten (Bharmal et al., 2010, J. Biomol Tech., 21 (3 Suppl): S43 supplied by Jena Bioscience GmbH and Thermo Fisher Scientific). Another type of method involves the introduction of a mutagenic nucleotide such as, but not limited to, 8-oxo-dGTP and dPTP (supplied from Jena Bioscience GmbH).

[00116] Predefined conditions are typically used for a cell, but not for isolated nuclei; however, labeling of nucleic acids as they are synthesized can be carried out using cells or nuclei isolated from cells.

[00117] In some embodiments of the invention, labeling can be applied to newly synthesized cDNA or DNA. Labeling can be used to identify a particular condition or subset of cells or nuclei. Thus, for example, different amounts of labels, such as nucleoside analog, hapten-labeled nucleotide, button-functionalized nucleotide and/or mutagenic nucleotide, and/or different ratios of labels can be used to specifically label an RNA, cDNA or DNA compartment. In another embodiment of the invention, a label may be added at various points in time to capture a point in time. Different labels or different ratios of labels can be added to differentially label RNA at different times. In some embodiments of the invention, labeling may be part of an indexing scheme for isolating individual cells. For example, a unique set of nucleotides for each compartment can be used in the extension step. Labeling can occur at a reverse transcription step, an extension step, a hybridization step, or an amplification step such as PCR. In some embodiments of the invention, this allows the detection of doublets or multiple cells or cell collisions.

Processing to obtain nucleic acid fragments

[00118] In one embodiment of the invention, the processing of isolated nuclei or cells can be used to fragment DNA nucleic acids in isolated nuclei or cells to obtain nucleic acid fragments (Fig. 1, block 14). Fragmentation of nucleic acids may be useful to obtain molecules, the length of which is suitable for sequencing methods described here. Processing may be necessary when the target nucleic acids to be sequenced are from DNA present in nuclei or cells; however, in some embodiments of the invention, processing is optional if the target nucleic acids to be sequenced are from RNA (e.g., from mRNA and/or non-coding RNA) present in nuclei or cells, since in some embodiments of the invention, the RNA molecules need not be be fragmented. In other embodiments of the invention, nucleic acids derived from RNA molecules are fragmented. Fragmentation can occur at any stage of this method. For example, the exemplary method shown in FIG. 2 includes fragmentation after adding two indices to nucleic acid molecules.

[00119] When processing nucleic acids in nuclei or cells, a nucleotide sequence is typically attached to one or both ends of the nucleic acid fragments resulting from the processing, and the nucleotide sequence can and usually does include one or more universal sequences. The universal sequence can be used, for example, as a "landing strip" in a subsequent step to anneal a nucleotide sequence, which can be used as a primer to add another nucleotide sequence, such as an index, to a nucleic acid fragment. The nucleotide sequence of such a primer may optionally include an index sequence. When processing nucleic acids in nuclei or cells, one or more unique molecular identifiers are usually added to one or both ends of the nucleic acid fragments resulting from the processing.

[00120] Various methods are known for processing nucleic acids in nuclei or cells to obtain nucleic acid fragments. Examples are CRISPR and Talen-like enzymes, as well as DNA unwinding enzymes (eg, helicases), which can form single-stranded regions to which DNA fragments can hybridize and initiate extension or amplification. For example, helicase-based amplification can be used (Vincent et al., 2004, EMBO Rep., 5(8):795-800). In one embodiment of the invention, extension or amplification is initiated with a randomized primer. In one embodiment of the invention, a transposome complex is used.

[00121] A transposome complex is a transposase associated with a transposase recognition site and can insert the transposase recognition site into a target nucleic acid within the nucleus in a process sometimes referred to as "tagging". At some of these insertion events, one strand of the transposase recognition site can be transferred to the target nucleic acid. Such a chain is called a "transferred chain". In one embodiment of the invention, the transposome complex comprises a dimeric transposase having two subunits and two non-contiguous transposon sequences. In another embodiment of the invention, the transposase includes a dimeric transposase having two subunits and a contiguous transposon sequence. In one embodiment of the invention, the 5' end of one or both strands of the transposase recognition site may be phosphorylated.

[00122] Some embodiments of the invention may include the use of an overactive Tn5 transposase and a Tn5 type transposase recognition site (Goryshin and Reznikoff, J. Biol. Chem., 273: 7367 (1998)) or a MuA transposase and a Mu transposase recognition site, containing R1 and R2 terminal sequences (Mizuuchi, K., Cell, 35: 785, 1983; Savilahti, H. et al., EMBO J., 14: 4893, 1995). Mosaic Tn5 (ME) end sequences can also be used as sequences optimized by one of skill in the art.

[00123] Additional examples of transposition systems that can be used with certain embodiments of the compositions and methods of the invention include Staphylococcus aureus Tn552 (Colegio et al., J. Bacteriol, 183: 2384-8, 2001; Kirby C et al, Mol. Microbiol ., 43: 173-86, 2002), Tyl (Devine & Boeke, Nucleic Acids Res., 22: 3765-72, 1994 and International Application Publication WO 95/23875), Tn7 transposon (Craig, NL, Science. 271: 1512, 1996; Craig, NL, Review in: Curr Top Microbiol Immunol., 204: 27-48, 1996), Tn/O and IS10 (Kleckner N, et al., Curr Top Microbiol Immunol., 204: 49-82 , 1996), Mariner transposase (Lampe DJ, et al., EMBO J., 15:5470-9, 1996); Tel (Plasterk RH, Curr. Topics Microbiol. Immunol., 204: 125-43, 1996); element P (Gloor, GB, Methods Mol. Biol., 260: 97-114, 2004); Tn3 (Ichikawa & Ohtsubo, J. Biol. Chem. 265: 18829-32, 1990); bacterial insertion sequences (Ohtsubo & Sekine, Curr. Top. Microbiol. Immunol. 204: 1-26, 1996); retroviruses (Brown et al., Proc Natl Acad Sci USA, 86:2525-9, 1989); and yeast retrotransposons (Boeke & Corces, Annu Rev Microbiol. 43: 403-34, 1989). Additional examples include IS5, Tn10, Tn903, IS911, and engineered variants of the transposase family of enzymes (Zhang et al., (2009) PIoS Genet. 5: e1000689. Epub 2009, Oct. 16; Wilson C. et al (2007) J. Microbiol Methods 71: 332-5).

[00124] Other examples of integrases that can be used in the methods and compositions described herein are retroviral integrases and integrase recognition sequences for such retroviral integrases, such as integrases derived from HIV-1, HIV-2, SIV, PFV-1, RSV.

[00125] The transposon sequences used in the methods and compositions described herein are provided in US Patent Application Publication No. 2012/0208705, US Patent Application Publication No. 2012/0208724, and International Patent Application Publication No. WO 2012/061832. In some embodiments, the transposon sequence includes a first transposase recognition site and a second transposase recognition site. In those embodiments where the transposome complex is used to introduce an index sequence, the index sequence may be present between transposase recognition sites or within a transposon.

[00126] Some of the transposome complexes used here include a transposase having two transposon sequences. In some such embodiments, the two transposon sequences are not linked to each other, that is, in other words, the transposon sequences are not contiguous. Examples of such transpos are known to those skilled in the art (see, for example, US Patent Application Publication No. 2010/0120098).

[00127] Typically, labeling is used to generate nucleic acid fragments that include a different nucleotide sequence at each end (eg, an N5 primer sequence at one end and an N7 primer sequence at the other end). This can be achieved by using two types of transposome complexes, where each transposome complex includes a different nucleotide sequence that is part of the transferred strand. In some embodiments of the invention, the labeling used here allows one nucleotide sequence to be inserted into nucleic acid fragments. Insertion of a nucleotide sequence results in nucleic acid fragments having a hairpin ligation duplex at one end and a nucleotide sequence into which the transposome complex is inserted at the other end. A nucleotide sequence comprising a transposome complex has a universal sequence. The universal sequence serves as a complementary sequence for hybridization in the amplification step described herein to introduce another index.

[00128] In some embodiments, the transposome complex includes a transposon nucleic acid sequence that links two transposase subunits to form a "loop complex" or "loop transposome". In one example, a transposome includes a dimeric transposase and a transposon sequence. Loop complexes can ensure the insertion of transposons into the target DNA while retaining information about the ordering of the original target DNA and without target DNA fragmentation. Clearly, loop structures can insert desired nucleic acid sequences, such as indices, into a target nucleic acid while maintaining a physical link to the target nucleic acid. In some embodiments, the transposon sequence of the loop transposome complex may include a fragmentation site such that the transposon sequence can be fragmented to form a transposome complex comprising two transposon sequences. Such transposome complexes can be used to ensure that adjacent target DNA fragments into which transposons are inserted will have barcode patterns that can be unambiguously assembled at a later stage of analysis.

[00129] In one embodiment of the invention, the fragmentation of nucleic acids is carried out using a fragmentation site present in the nucleic acids. Typically, fragmentation sites are introduced into target nucleic acids using a transposome complex. In one embodiment of the invention, after nucleic acid fragmentation, the transposase remains associated with nucleic acid fragments, and therefore, nucleic acid fragments derived from the same genomic DNA molecule remain physically associated (Adey et al., 2014, Genome Res. , 24:2041-2049). For example, a loop transposome complex may include a fragmentation site. The fragmentation site can be used for physical cleavage, but not informational association, between index sequences that have been inserted into the target nucleic acid. Cleavage can be carried out by biochemical, chemical or other means. In some embodiments of the invention, the site of fragmentation may include a nucleotide or nucleotide sequence that can be fragmented in various ways. Examples of fragmentation sites include, but are not limited to, a restriction endonuclease site, at least one ribonucleotide cleaved by an RNase; nucleotide analogs cleavable in the presence of a specific chemical agent; a diol bond cleaved by treatment with periodate; a disulfide group cleavable by a chemical reducing agent; a cleavable fragment that can be subjected to photochemical cleavage; and a peptide cleavable by a peptidase enzyme or other suitable methods (see, for example, US Patent Application Publication No. 2012/0208705, US Patent Application Publication No. 2012/0208724 and WO 2012/061832).

[00130] The transposome complex may optionally include an index sequence, also referred to as a transposase index. The index sequence is present as part of the transposon sequence. In one embodiment of the invention, the index sequence may be present on the transferred strand, ie, the strand of the transposase recognition site, which is transferred to the target nucleic acid.

[00131] Following nuclear labeling and processing of nucleic acid fragments, a purification step may be performed to increase the purity of the molecules. Any suitable purification method, such as electrophoresis, size exclusion chromatography, or the like, may be carried out. In some embodiments of the invention, paramagnetic spheres for solid-phase reversible immobilization can be used to separate the desired DNA molecules, for example, from non-included primers and to select nucleic acids for size. Paramagnetic spheres for solid phase reversible immobilization are commercially available from Beckman Coulter (Agencourt AMPure XP), Thermofisher (MagJet), Omega Biotek (Mag-Bind), Promega Beads (Promega), and Kara Biosystems (Kara Pure Beads).

Adding a Compartment-Specific Index

[00132] An index sequence, also referred to as a label or barcode, is suitable as a marker for characterizing the compartment in which a particular nucleic acid is present. Accordingly, an index is a sequence of a tag nucleic acid that is attached to each target nucleic acid present in a particular compartment, and the presence of this tag indicates a compartment or allows identification of a compartment in which a population of isolated nuclei or cells is present on a particular compartment. steps of this method. The addition of an index to nucleic acid fragments is carried out using subpopulations of isolated nuclei or cells distributed in different compartments (Fig. 1, block 15; Fig. 2, blocks 24, 26 and 30; Fig. 3, blocks 33 and 37).

[00133] The index sequence may be any suitable number of nucleotides in length, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more nucleotides. A four-base label makes it possible to determine the multiplexity of 256 samples in one array, and a six-base label allows processing 4096 samples in one array.

[00134] In one embodiment of the invention, the addition of the index is achieved during the processing of nucleic acids to obtain nucleic acid fragments. Thus, for example, a transposome complex can be used that includes an index. In some embodiments of the invention, the index is added after processing of nucleic acid fragments containing the nucleotide sequence at one or both ends. In other embodiments of the invention, no processing is required to add the index. Thus, for example, an index can be added directly to RNA nucleic acids without fragmenting the RNA nucleic acids. Accordingly, the term "nucleic acid fragment" includes nucleic acids that result from processing and RNA nucleic acids, as well as nucleic acids derived from these nucleic acids.

[00135] Methods for adding an index include, but are not limited to, ligation, extension (including extension using reverse transcriptase), hybridization, adsorption, specific or non-specific primer interactions, amplification, or transposition. The nucleotide sequence that is added to one or both ends of the nucleic acid fragments may also include one or more universal sequences and/or unique molecular identifiers. A universal sequence can be used, for example, as a "landing strip" in a subsequent step to anneal a nucleotide sequence, which can be used as a primer to add another nucleotide sequence, such as another index and/or another universal sequence, to a nucleic acid fragment. . Thus, a method that includes one, two, or more steps can be used to incorporate an index sequence using essentially any combination of ligation, extension, hybridization, adsorption, specific or non-specific primer interactions, amplification, or transposition.

[00136] For example, in embodiments of the invention that use nucleic acid fragments derived from mRNA, various methods can be applied to add an index to the mRNA in one or two steps. For example, the index can be added using cDNA production techniques. A primer with a poly-T sequence at the 3' end can be hybridized to mRNA molecules and extended using reverse transcriptase. Treatment of isolated nuclei or cells with these components under conditions suitable for reverse transcription allows one-step index addition, which will result in a population of indexed nuclei or cells, where each nucleus or cell contains indexed nucleic acid fragments. Alternatively, a primer with a poly-T sequence includes a universal sequence instead of an index, and the index is added in a subsequent step of ligation, primer extension, amplification, hybridization, or a combination thereof. In some embodiments of the invention, the barcode is added without using a universal sequence. The indexed nucleic acid fragments in the synthesized strand may and usually include an index sequence indicating a particular compartment.

[00137] In embodiments of the invention that involve the use of nucleic acid fragments derived from non-coding RNA, various methods of adding an index to non-coding RNA in one or two steps can be applied. Thus, for example, an index can be added using a first primer that includes a randomized sequence and a matrix switch primer, where either primer can include an index. A reverse transcriptase having terminal transferase activity that results in the addition of non-template nucleotides to the 3' end of the synthesized strand can be used, and the template-switching primer includes nucleotides that hybridize to non-template nucleotides added by reverse transcriptase. An example of a suitable reverse transcriptase enzyme is Moloney's murine leukemia virus reverse transcriptase. In a specific embodiment of the invention, the SMARTer™ reagent available from Takara Bio USA, Inc. (Cat. No. 634926) is used to switch the template in order to add an index to non-coding RNA and mRNA, if desired. Alternatively, the first primer and/or template switch primer may include a universal sequence instead of an index, and the index is added in a subsequent step of ligation, primer extension, amplification, hybridization, or a combination thereof. The indexed nucleic acid fragments on the synthesized strand may and usually include an index sequence indicating a particular compartment. Other embodiments of the invention include 5' or 3' RNA profiling or full length RNA profiling.

[00138] In another embodiment of the invention, specific mRNA and/or non-coding RNA can be targets for amplification. Targeting enables the creation of sequencing libraries enriched with sequences that are more likely to provide useful information, thereby significantly reducing sequencing depth and associated costs, and increasing the ability to detect subtle differences between cells. RNA molecules comprising one or more mRNAs and/or one or more non-coding RNAs can be selected for the likelihood of obtaining useful information, and primers can be used to selectively anneal predetermined RNA nucleic acids and amplify a subset of total RNA molecules present in a cell or in the core. One skilled in the art will appreciate that the appropriate RNA molecules to select depend on the experiment. Thus, for example, when evaluating non-coding perturbations, only the coding regions that are in the cis-position with respect to the disrupted regulatory element can be tested for a change in the level of expression. This approach reduces the background of ribosomal reads to a greater extent than the use of a randomized hexamer or poly-T primers. This approach also allows targeting of splicing junctions and exons formed as a result of alternative events at the transcription initiation site, and thus obtain isoform information that is difficult to obtain with standard sci methods.

[00139] Targeted amplification of RNA molecules can occur in several steps during library preparation. In one embodiment of the invention, targeted amplification of multiple targets occurs during reverse transcription of RNA molecules. Many different primers targeting different RNA molecules can be used in an experiment. In one embodiment of the invention, many different primers can be used to target different regions of the same RNA molecule. The use of multiple primers targeting different regions of the same RNA molecule provides multiple opportunities for reverse transcription of the RNA molecule into cDNA, which will increase the likelihood of detecting the RNA molecule.

[00140] In one embodiment of the invention, the primers used for targeted amplification do not include an index. If the index is not added during the amplification reaction, then the allocation of cells or nuclei to different compartments is optional, and amplification can occur in one reaction with all RNA molecules and all primers present. In those embodiments where the index is added during the amplification reaction, the distribution of cells or nuclei is beneficial and amplification can be carried out in one reaction in each compartment in the presence of all RNA molecules and all primers, but, at the same time, each primer in the compartment has the same compartment-specific index.

[00141] In one embodiment of the invention, the design of primers for multiplex target capture may be based on one or more of the following considerations. Once the RNA has been selected for targeted amplification, the RNA sequence can be assembled and all possible reverse transcriptase primers, ie, candidate primers, can be determined. The length of any primer should be sufficient for their function in the reverse transcription reaction, and may be, for example, from 20 to 30 nucleotides.

[00142] Candidate primers can be filtered by various criteria including, but not limited to, GC content, localization of GC bases in the primer, likelihood of targeting an unwanted site, and mapability. A suitable GC content is 40-60%, which corresponds to melting points that are approximately 55-70°C. Preferably, these two guanine or cytosine bases are present in the last 5 nucleotides at the 3' end of the primer, which will increase the likelihood that the hybridized primer will be a good substrate for extension by the reverse transcriptase enzyme.

[00143] With regard to the likelihood of unwanted priming, the authors of the present invention found that although the target RNAs were highly enriched, however, most of the reads were still derived from other RNAs that were present in cells in large quantities. Most of these undesired priming events result from approximately 5-8 base pairs being complementary between the 3' end of the primer and the non-target RNA. The authors of the present invention have found that the redundancy of the final hexamer of the candidate primer in the total cellular RNA should be taken into account. Suitable primers have been determined to include the last hexamer which is either (i) not present in ribosomal RNA or (ii) present at low levels in total cellular RNA.

[00144] Examples of hexamers not present in ribosomal RNA are described (hexamers that "have not been appropriately randomized" or NSR hexamers are described by Armor et al., 2009, Nature Methods, 6(9): 647-49). It has been found that primers with this property are much less likely to have off-target priming in ribosomal RNA. One method for determining whether a hexamer is present at a low level in total cellular RNA is a method that may involve identifying an excess of each hexamer in the RNA molecules in the cell, e.g., increasing transcription, including ribosomal transcription, in cells of the analyzed type according to with the methods described here. The use of candidate primers that have a slight excess, eg, within the lowest quartile, may reduce targeting to an undesirable site.

[00145] Candidate primers can also be assessed for mapability. So, for example, each candidate can be aligned to the targets using a bow tie algorithm and allows 3 mismatches. This step will ensure that each primer has only one target site in the genome.

[00146] In some embodiments, the amplification of multiple targets in the same reaction, also referred to as multiplex target capture, with regulation of reverse transcriptase primer anneal temperatures, can be used to maintain specific reverse transcription and amplify desired target RNAs. For example, typical reverse transcription protocols include denaturing a mixture of RNA and reverse transcription primer and cooling to 4°C to ensure annealing. The low annealing temperature creates conditions of too low stiffness and results in undesired annealing events. To increase the likelihood that only annealing events will be observed when all reverse transcription target primers hybridize to the correct targets, a high temperature must be maintained throughout the entire reverse transcription process. In one embodiment of the invention, components, e.g., a mixture of fixed cells, a reverse transcription primer pool and dNTP at 65°C, hybridized at 53°C, are added to a reverse transcription enzyme/buffer mixture that has been pre-equilibrated at 53°C to annealing reaction, and then carry out the elongation reaction at 53°C for 20 minutes. Thus, the likelihood of annealing reverse transcription primers at low temperature between the denaturation and extension steps is reduced. One skilled in the art will appreciate that some changes can be made, such as changing temperature or time, without reducing reverse transcription specificity.

[00147] Other methods can be used to add an index to a nucleic acid fragment, and the method for adding an index is not particularly limited. Thus, for example, in one embodiment of the invention, the introduction of the index sequence includes ligating the primer to one or both ends of the nucleic acid fragments. Ligation of the ligation primer may be facilitated by the presence of a universal sequence at the ends of the nucleic acid fragments. An example of a primer is a hairpin ligation duplex. The ligation duplex may be ligated to one end, or preferably to both ends, of the nucleic acid fragments.

[00148] In another embodiment of the invention, the introduction of the index sequence includes the use of single-stranded nucleic acid fragments and the synthesis of the second DNA strand. In one embodiment of the invention, the second DNA strand is prepared using a primer that includes sequences that are complementary to the nucleotides present at the ends of the single stranded nucleic acid fragments.

[00149] In another embodiment of the invention, index insertion occurs in one, two, three, or more rounds of split-and-merge barcoding, resulting in single cell libraries with one, two, three, or multiple (e.g., four or more) indexes.

[00150] In another embodiment of the invention, the inclusion of indices and an amplification mediator (eg, a universal sequence) is useful, and allows the production of single cell target sequencing libraries and/or single cell target sequencing libraries.

Addition of universal sequences for immobilization

In one embodiment of the invention, the addition of nucleotides during the processing and/or indexing steps results in the addition of universal sequences suitable for fragment immobilization and sequencing. In another embodiment of the invention, indexed nucleic acid fragments can also be processed to add universal sequences suitable for immobilization and sequencing of nucleic acid fragments. One skilled in the art will appreciate that, in those embodiments where the compartment is a droplet, sequences for immobilizing nucleic acid fragments are optional. In one embodiment of the invention, the introduction of universal sequences suitable for immobilization and sequencing of fragments includes the ligation of identical universal adapters (also referred to as "mismatched adapters", the general features of which are described by Gormley et al., in US patent 7741463 and Bignell et al. in patent US 8053192) with the 5'-3' ends of the indexed nucleic acid fragments. In one embodiment of the invention, the universal adapter includes all sequences required for sequencing, including sequences for immobilizing indexed nucleic acid fragments on an array.

[00152] In one embodiment of the invention, blunt end ligation may be used. In another embodiment of the invention, the nucleic acid fragments are prepared so that they have single overhanging nucleotides, for example, by the activity of certain types of DNA polymerase, such as Taq polymerase or exo-(-)-Klenow polymerase, which has an independent matrix terminal transferase activity adding one or more deoxynucleotides, for example deoxyadenosine (A) to the 3' ends of the indexed nucleic acid fragments. In some cases, the overhanging nucleotide is more than one base. Such enzymes can be used to add one "A" nucleotide to the blunt end of the 3' end of each strand of nucleic acid fragments. Thus, "A" can be added to the 3' end of each strand of double-stranded target fragments by reaction with Taq polymerase or Klenow exo-(-) polymerase, and additional sequences added to each end of the nucleic acid fragment can include a compatible "T" overhang present at the 3' end of each region of the added double-stranded nucleic acid. This terminal modification also prevents the autoligation of nucleic acids, and therefore there is a bias towards the formation of indexable nucleic acid fragments flanked by the sequences that are added in this embodiment of the invention.

[00153] In another embodiment of the invention, if the universal adapter ligated to the indexed nucleic acid fragments does not include all the sequences required for sequencing, then an amplification step, such as PCR, can be performed to further modify the universal adapters present in each indexed nucleic acid fragment prior to immobilization and sequencing. Thus, for example, the initial primer extension reaction can be performed using a universal anchor sequence that is complementary to the universal sequence present in the indexed nucleic acid fragment that produces extension products that are complementary to both strands of each individual indexed nucleic acid fragment. Typically, PCR adds additional universal sequences, such as a universal capture sequence.

[00154] After adding universal adapters, either by a one-step ligation method, or by hybridization of a universal adapter that includes all sequences required for sequencing, or by a two-step method of ligating a universal adapter, and then amplifying to further modify the universal adapter, the final index fragments will include a universal sequence to capture and anchor sequence. The result of adding universal adapters to each end is a set or library of indexed nucleic acid fragments.

[00155] The resulting indexed fragments, taken together, are a library of nucleic acids that can be immobilized and then sequenced. The term "library", also referred to here as a "sequencing library", means a set of nucleic acid fragments from individual nuclei or cells containing known universal sequences at their 3' and 5' ends. This library includes nucleic acids from the entire transcriptome, nucleic acids from newly synthesized RNA molecules, or a combination thereof, and can be used to sequence the entire transcriptome, the transcriptome of newly synthesized RNA, or a combination of both.

[00156] Indexed nucleic acid fragments can be processed under conditions that allow predetermined size molecules to be selected, eg 150 to 400 nucleotides in length, eg 150 to 300 nucleotides. The resulting indexed nucleic acid fragments are combined and then optionally purified to increase the purity of the DNA molecules by removing at least a portion of non-included universal adapters or primers. Any suitable purification method, such as electrophoresis, size exclusion chromatography, or the like, may be carried out. In some embodiments of the invention, paramagnetic spheres for solid-phase reversible immobilization can be used to separate the desired DNA molecules, for example, from non-included primers, and to select nucleic acids for size. Paramagnetic spheres for solid phase reversible immobilization are commercially available from Beckman Coulter (Agencourt AMPure XP), Thermofisher (MagJet), Omega Biotek (Mag-Bind), Promega Beads (Promega), and Kapa Biosystems (Kapa Pure Beads).

[00157] A non-limiting illustrative embodiment of the invention is shown in FIG. 1. In this embodiment of the invention, the method includes obtaining a plurality of cells (Fig. 1, block 10). This method also includes the distribution of subpopulations of cells across multiple compartments (Fig. 1, block 11) and treatment of cells under predetermined conditions (Fig. 1, block 12). The predetermined conditions may and usually differ between different compartments. Thus, for example, different compartments may include different doses of the agent, different perturbations, different treatment times, different cell types, and the like. The newly synthesized RNA is then labeled (Fig. 1, block 13). Cells can be labeled, or nuclei can be isolated from cells and then labeled. In some embodiments of the invention, cellular RNA is processed into fragments (FIG. 1, block 14). RNA present in cells or nuclei is then indexed (FIG. 1, block 15). In this case, various variants of RNA indexing are possible. Thus, for example, in one embodiment of the invention, all mRNAs present in the cell are indexed using a primer that includes a poly-T region. In another embodiment of the invention, specific RNA nucleic acids can be indexed.

[00158] Another non-limiting illustrative embodiment of the invention is shown in FIG. 2 and described in Example 1. In this embodiment, the method includes obtaining isolated nuclei from a plurality of cells (FIG. 2, block 22). This method also includes the distribution of subpopulations of selected nuclei across the first set of compartments (Fig. 2, block 23). The number of compartments in the first distribution stage (FIG. 2, block 23) may depend on the format used. For example, the number of compartments can be from 2 to 96 compartments (using a 96-well plate), from 2 to 384 compartments (using a 384-well plate), or from 2 to 1536 compartments (using a 1536-well plate). Alternatively, other compartments such as drops can be used.

[00159] The method also includes obtaining indexable cores (FIG. 2, block 24). In one embodiment of the invention, obtaining indexed nuclei includes using a reverse transcriptase with an oligo-dT primer to add an index, a randomized nucleotide sequence, and a universal sequence. The index within each compartment is unique, eg each index is compartment-specific. The randomized sequence is used as a unique molecular identifier (UMI) to label unique nucleic acid fragments. A randomized sequence can also be used to facilitate the removal of duplicates in post-processing. The universal sequence serves as a complementary sequence for hybridization in the ligation step described here. In another embodiment of the invention, obtaining indexed nuclei includes the use of reverse transcriptase with specific primers to target predetermined RNA molecules. Reverse transcription may result in the addition of an index, a randomized nucleotide sequence, and a universal sequence to target RNA molecules. Treatment of the nuclei with these components under conditions suitable for reverse transcription results in the formation of a population of indexed nuclei, where each nucleus contains indexed nucleic acid fragments. Indexed nucleic acid fragments may, and usually include, on the synthesized strand, an index sequence indicating a particular compartment. An example of an indexed nucleic acid fragment is shown in FIG. 1A of Example 1 (see Indexed Reverse Transcription section).

[00160] Indexed nuclei from multiple compartments can be combined (FIG. 2, block 25). Subpopulations of these pooled indexed nuclei, referred to herein as pooled indexed nuclei, are then distributed across a second set of compartments (FIG. 2, block 25). After distributing the nuclei into subpopulations, a second index sequence is included in each compartment of the indexed nucleic acid fragments to obtain fragments with two indexes. This results in further indexing of the indexed nucleic acid fragments (FIG. 2, block 26).

[00161] In this illustrative embodiment of the invention, the introduction of the second index sequence includes duplex ligation, ligation of the hairpin ligation duplex with indexed nucleic acid fragments in each compartment. The use of a hairpin ligation duplex to introduce a universal sequence, index, or combination thereof at the end of a target nucleic acid fragment typically involves using one end of the duplex as a primer for subsequent amplification. In contrast, the hairpin ligation duplex used in this embodiment does not act as a primer. The advantage of using the hairpin ligation duplex described here is the reduction in autoligation observed with many hairpin ligation duplexes described in the literature. In one embodiment, the ligation duplex includes five elements: 1) a universal sequence that is complementary to the universal sequence present in the oligo-dT primer, 2) a second index, 3) i-deoxy-U, 4) a nucleotide sequence that can form a hairpin and 5) reverse complement of the second index. The second index sequences are unique for each compartment into which distributed indexed nuclei were placed (FIG. 2, block 25) after adding the first index by reverse transcription. An example of a nucleic acid fragment with two indices is shown in FIG. 1A of Example 1 (see "Indexed hairpin ligation").

[00162] Removal of i-deoxy-U present in the hairpin region of the hairpin ligation duplex incorporated into nucleic acid fragments can occur before, during or after purification. Removal of the uracil residue can be accomplished by any available means, and in one embodiment, the Uracil Specific Excision Reagent (USER) supplied by NEB is used.

[00163] Subpopulations of these two-index pooled nuclei, referred to herein as two-index pooled nuclei, are then distributed to a third set of compartments (FIG. 2, block 27). In one embodiment of the invention, 100 to 30,000 cores are dispensed into each well. In one embodiment, the number of cores per well is at least 100, at least 500, at least 1000, or at least 5000. In one embodiment, the number of cores per well is at most 30,000, at most 25,000, not more than 20000 or not more than 15000. In one embodiment of the invention, the number of nuclei present in the subpopulation may be 100-1000, 1000-10000, 10000-20000 or 20000-30000. In one embodiment of the invention, 2500 cores are placed in each well. In one embodiment of the invention, the number of nuclei present in each subpopulation is approximately the same.

[00164] After the distribution of nuclei with two indices to subpopulations, the synthesis of the second strand of DNA is carried out (Fig. 2, block 28). Nucleic acids in the nuclei are processed by labeling (FIG. 2, block 29). Each compartment containing nuclei with two indexes includes a transposome complex. In this embodiment, labeling is performed to produce nucleic acid fragments that include a different nucleotide sequence at each end (eg, an N5 primer sequence at one end and an N7 primer sequence at the other end).

[00165] After nuclear labeling, nucleic acid fragments with two indices in each compartment are introduced with a third index sequence to obtain fragments with three indices, where the third index sequence in each compartment differs from the first and second index sequences in the compartments. This results in additional indexing of indexed nucleic acid fragments (FIG. 2, block 30; see also FIG. 1A of the example ("PCR-indexed USER treatment")) prior to immobilization and sequencing. In one embodiment of the invention, universal sequences present at the ends of nucleic acid fragments with two indexes (for example, a nucleotide sequence inserted into a hairpin ligation duplex at one end and a nucleotide sequence inserted into a transposome complex at the other end) can be used to primer binding and for extension in the amplification reaction. Typically, two different primers are used. One primer hybridizes to universal sequences at the 3'end of one strand of double-index nucleic acid fragments, and the second primer hybridizes to universal sequences at the 3'end of the other strand of double-index nucleic acid fragments. Thus, the anchor sequences (eg, the site to which a universal primer hybridizes, such as a read 1 or read 2 sequencing primer) present on each primer may differ. Each of the suitable primers may include additional universal sequences, such as a universal capture sequence (eg, a site to which the capture oligonucleotide hybridizes, where the capture oligonucleotide can be immobilized on the surface of a solid support). Since each primer includes an index, this step results in the addition of one more index sequence, one at each end of the nucleic acid fragments, resulting in fragments with three indexes. In one embodiment of the invention, indexed primers such as indexed primer P5 and indexed primer P7 can be used to add a third index. Fragments with three indexes are combined, and then they can be subjected to the stage of purification described here.

[00166] A plurality of fragments with three indexes can be obtained for sequencing. After combining fragments with three indexes and after their purification, they are enriched, usually by immobilization and/or amplification before sequencing (Fig. 2, block 31).

[00167] Another non-limiting illustrative embodiment of the invention is shown in FIG. 3 and described in Example 4. In this embodiment, the method includes obtaining isolated nuclei or cells (FIG. 3, block 30). Nuclei or cells can be treated with reverse transcriptase and specific primers to target and enrich predetermined RNA molecules (FIG. 3, block 31). Treatment of nuclei or cells with these components under conditions suitable for reverse transcription produces a population of nuclei or cells, where each nucleus contains nucleic acid fragments enriched in sequences present in predetermined RNA molecules. This method also includes the distribution of subpopulations of nuclei or cells in the first set of compartments (Fig. 3, block 32). The number of compartments in the first distribution stage (FIG. 3, block 32) may depend on the format used. For example, the number of compartments can be from 2 to 96 compartments (using a 96-well plate), from 2 to 384 compartments (using a 384-well plate), or from 2 to 1536 compartments (using a 1536-well plate). Alternatively, other compartments such as drops can be used.

[00168] This method also includes obtaining indexed nuclei or cells (Fig. 3, block 33) by incorporating nucleic acid fragments in each compartment of the index sequence to obtain indexed fragments.

[00169] In one embodiment of the invention, the introduction of the index sequence includes ligation of the hairpin ligation duplex with indexed nucleic acid fragments in each compartment. The nuclei or cells containing the indexed fragments are pooled and then subpopulations of these pooled indexed nuclei or cells are distributed into a second set of compartments (FIG. 3, block 34).

[00170] The distribution of indexed nuclei or cells into subpopulations may be accompanied by the synthesis of a second strand of DNA (Fig. 3, block 35). Nucleic acids in nuclei or cells are processed by labeling (FIG. 3, block 36). Each compartment containing indexed nuclei includes a transposome complex. In this embodiment, labeling is performed to produce nucleic acid fragments that include a different nucleotide sequence at each end (eg, an N5 primer sequence at one end and an N7 primer sequence at the other end).

[00171] After labeling the nuclei, a second index sequence is introduced into the nucleic acid fragments with two indices in each compartment to obtain fragments with two indices, where the second index sequence in each compartment differs from the first index sequences in the compartments. This results in additional indexing of the indexed nucleic acid fragments (FIG. 3, block 37) prior to immobilization and sequencing.

[00172] For sequencing, a plurality of fragments with two indices can be obtained, where the sequencing data is enriched with sequences present in predetermined RNA molecules. After pooling and purification of fragments with two indexes, they are usually enriched by immobilization and/or amplification before sequencing (Fig. 3, block 38).

Obtaining immobilized samples for sequencing

[00173] Methods for attaching indexed fragments from one or more sources to a substrate are known to those skilled in the art. In one embodiment of the invention, the indexed fragments are enriched using a plurality of capture oligonucleotides specific to the indexed fragments, and the capture oligonucleotides can be immobilized on the surface of a solid support. For example, the capture oligonucleotides may include a first member of the universal binding pair, and the second member of the binding pair may be immobilized on the surface of a solid support. Similarly, methods for amplifying dual-index immobilized fragments include, but are not limited to, bridge amplification and kinetic exclusion. Methods for immobilization and amplification prior to sequencing are described, for example, by Bignell et al. (US patent 8053192), Gunderson et al. (WO 2016/130704), Shen et al. (US patent 8895249) and Pipenburg et al. (U.S. Patent 9309502).

[00174] The pooled sample may be immobilized in preparation for sequencing. Sequencing can be performed using an array of individual molecules, or these molecules can be amplified prior to sequencing. Amplification can be carried out using one or more immobilized primers. The immobilized primer(s) may be, for example, a track on a flat surface or on a pool of spheres. A pool of spheres can be isolated to form an emulsion with one sphere in each "compartment" of the emulsion. By concentrating only one template per "compartment", only one template is amplified per sphere.

[00175] As used herein, the term "solid-phase amplification" means any nucleic acid amplification reaction carried out on a solid support or in combination with a solid support, such that all or part of the amplified products are immobilized on the solid support as they are formed. In particular, the term encompasses solid phase polymerase chain reaction (solid phase PCR) and solid phase isothermal amplification, which are reactions similar to standard liquid phase amplification, except that one or both of the forward and reverse amplification primers are immobilized on a solid support. Solid phase PCR covers systems such as emulsions, in which one primer is anchored to a sphere and another is present in free solution, and colony formation in solid phase gel matrices, where one primer is anchored to the surface and the other is in free solution.

[00176] In some embodiments of the invention, the solid media contains a profiled surface. The term "shaped surface" refers to the location of various areas within or on the exposed layer of the solid carrier. Thus, for example, one or more regions may be elements in which one or more amplification primers are present. These elements can be separated by intermediate regions where there are no amplification primers. In some embodiments of the invention, the structure may be elements in x-y format that are in rows and columns. In some embodiments of the invention, the structure may be a repeating arrangement of elements and/or intermediate areas. In some embodiments of the invention, the structure may be a random arrangement of elements and/or intermediate areas. Representative shaped surfaces that can be used in the methods and compositions described herein are those described in U.S. Patent No. 8778848, 8778849 and 9079148 and U.S. Patent Application Publication No. 2014/0243224.

[00177] In some embodiments of the invention, the solid media includes an array of holes or depressions in the surface. This can be obtained by generally known methods using a number of technologies, including, but not limited to, photolithography, stamping methods, molding methods, and microetching methods. As will be understood by those skilled in the art, the technology employed will depend on the composition and shape of the substrate array.

[00178] The features in the shaped surface may be wells in an array of wells (e.g., microwells or nanowells) on glass, silicon, plastic, or other suitable solid supports with a structured, covalently bonded gel, such as a copolymer (N-(5-azidoacetamidylpentyl) acrylamide and acrylamide) (PAZAM, see for example US Patent Application Publication No. 2013/184796, WO 2016/066586 and WO 2015/002813). This method allows the creation of gel pads used for sequencing that can be stable over high cycle sequencing rounds. The covalent binding of the polymer to the wells allows the retention of the gel with structured properties throughout the half-life of the structured support in a variety of applications. However, in many embodiments of the invention, the gel need not be covalently bonded to the wells. For example, under certain conditions, silane-free acrylamide (SFA, see, for example, US Pat. No. 8,563,677) that is not covalently bound to any part of the structured support can be used as a gel.

[00179] In particular embodiments, a structured substrate can be made by patterning a solid support with wells (e.g., microwells or nanowells) by coating the structured support with a gel material (e.g., PAZAM, SFA, or chemically modified variants thereof, such as azidolized a variant of SFA (azido-SFA) and polishing of the gel-coated carrier, for example by chemical or mechanical polishing, retaining the gel in the wells, but removing or inactivating nearly all of the gel from the intermediate regions on the surface of the structured carrier between the wells. The solution of the indexed fragments can then be contacted with a polished substrate such that the individual indexed fragments are seeded into individual wells through interactions with primers bound to the gel material, however, the nuclein The target acids will not occupy the intermediate regions due to the lack of gel material or its lack of activity. Amplification of the indexed fragments will be limited to wells because the lack of gel or the lack of gel activity in the intervening regions prevents outward migration of the growing nucleic acid colony. This method can be easily carried out on an industrial scale, is scalable and involves the use of standard micro- or nanomachining techniques.

[00180] Although the disclosure of the present invention covers methods of "solid-phase" amplification, where only one amplification primer was immobilized (the other primer is usually present in free solution), however, in one embodiment of the invention, it is preferable that the solid support be provided with forward and reverse immobilized primers. In practice, there will be a "multiple" identical forward primers on a solid support and/or a "multiple" identical reverse primers immobilized on a solid support, since the amplification process requires an excess of primers to maintain amplification. References to forward and reverse primers in the specification of this application are accordingly interpreted to encompass a "multiple" of such primers, unless this is contrary to the context of the invention.

[00181] As will be appreciated by one of skill in the art, any given amplification reaction requires the presence of at least one type of forward primer and at least one type of reverse primer specific to the template being amplified. However, in certain embodiments of the invention, the forward and reverse primers may include template-specific portions of an identical sequence and may have completely identical nucleotide sequences and structures (including any non-nucleotide modifications). In other words, it is possible to carry out solid phase amplification using only one type of primer, and such single primer methods are within the scope of the present invention. In other embodiments of the invention, forward and reverse primers can be used that contain identical template-specific sequences but that differ in some other structural features. Thus, for example, one type of primer may contain a non-nucleotide modification that is not present in another primer.

[00182] In all embodiments of the disclosure, primers for solid phase amplification are preferably immobilized by covalently linking in the same position with a solid support at the 5' end or near the 5' end of the primer, which allows the template-specific portion of the primer to freely hybridize to its cognate template and with a 3'-hydroxyl group free to extend the primer. For this purpose, any suitable method of covalent bonding known to those skilled in the art may be used. The chemical bonding method chosen will depend on the nature of the solid support and any derivatization or functionalization thereof. The primer itself may include a moiety, which may be a non-nucleotide chemical modification, to facilitate binding. In a specific embodiment of the invention, the primer may include a sulfur-containing nucleophile, such as phosphorothioate or thiophosphate, at the 5' end. In the case of solid-supported polyacrylamide hydrogels, this nucleophile will bind to the bromoacetamide group present in the hydrogel. A more specific method of coupling primers and templates to a solid support is the 5'-phosphorothioate coupling method to a hydrogel consisting of polymerized acrylamide and N-(5-bromoacetamidylpentyl)acrylamide (BRAPA) as described in WO 05/065814.

[00183] In some embodiments, solid carriers may be used that include an inert support or matrix (e.g., glass slides, polymer spheres, etc.) that has been "functionalized," e.g., by deposition of a layer or coating of an intermediate , which includes reactive groups that provide covalent binding to biomolecules such as polynucleotides. Examples of such carriers are, but are not limited to, polyacrylamide hydrogels supported on an inert support such as glass. In such embodiments, the biomolecules (eg, polynucleotides) may be directly covalently linked to an intermediate (eg, a hydrogel), but the intermediate itself may not be covalently linked to a backing or matrix (eg, a glass base). The term "covalent bonding to a solid support" should be interpreted, accordingly, as a term covering this type of structure.

[00184] Pooled samples can be amplified on spheres, where each sphere contains a forward and reverse amplification primer. In a specific embodiment of the invention, a library of indexed fragments is used to obtain clustered arrays of nucleic acid groups similar to those described in US patent application publication No. 2005/0100900, in US patent No. 7115400, in US patent application WO 00/18957 and WO 98/44151, by solid phase amplification, and more specifically, solid phase isothermal amplification. The terms "cluster" and "group" as used herein are synonymous and refer to a discrete site on a solid support comprising a plurality of identical immobilized nucleic acid strands and a plurality of identical immobilized complementary nucleic acid strands. The term "clustered array" means an array formed by such clusters or groups. In this context, the term "array" should not be understood as an ordered arrangement of clusters.

[00185] As used herein, the term "solid phase" or "surface" means any flat array where primers are attached to a flat surface, such as glass, silicon dioxide, or plastic microscope slides or similar flow cell devices; spheres where one or two primers are attached to the spheres and the spheres are amplified; or an array of spheres on the surface after sphere amplification.

[00186] Clustered arrays can be obtained using any thermocell reaction method as described in WO 98/44151, or a method in which a constant temperature is maintained and the elongation and denaturation cycles are performed by changing reagents. Such isothermal amplification methods are described in WO No. 02/46456 and US Patent Application Publication 2008/0009420. Due to the lower temperatures used in the isothermal process, this method is particularly preferred in some embodiments of the invention.

[00187] It should be noted that any amplification methods described in this application or, as such, known to specialists in this field, can be carried out using universal or target-specific primers to amplify immobilized DNA fragments. Suitable amplification methods include, but are not limited to, polymerase chain reaction (PCR), strand change amplification (SDA), transcription-mediated amplification (TMA), and nucleic acid sequence-based amplification (NASBA), as described in U.S. Patent No. No. 8003354. The above amplification methods can be used to amplify one or more nucleic acids of interest. Thus, for example, PCR, including multiplex PCR, SDA, TMA, NASBA, and the like. can be used to amplify immobilized DNA fragments. In some embodiments of the invention, primers that are specific for the polynucleotide of interest are included in the amplification reaction.

[00188] Other suitable polynucleotide amplification methods may include oligonucleotide extension and ligation methods, rolling ring amplification (RCA) (Lizardi et al., Nat. Genet. 19: 225-232 (1998)) and oligonucleotide ligation assay (OLA) (See, in general terms, US Pat. Obviously, these amplification techniques can be developed to amplify immobilized DNA fragments. For example, in some embodiments, the amplification method may include amplification reactions for probe ligation or oligonucleotide ligation assay (OLA) that include primers that specifically target the nucleic acid of interest. In some embodiments, the amplification method may include a primer ligation-extension reaction that includes primers specifically targeted to the nucleic acid of interest. As a non-limiting example of primers for extension and ligation, primers can be specifically designed to amplify the nucleic acid of interest, and such amplification can include primers used for the GoldenGate assay (Illumina, Inc., San Diego, CA) described in patents. USA No. 7582420 and 7611869.

[00189] DNA nanobeads can also be used in combination with the methods and compositions described herein. Methods for obtaining and using DNA nanobeads for genomic sequencing can be found, for example, in US patents and US patent publications 7910354, in 2009/0264299, 2009/0011943, 2009/0005252, 2009/0155781, 2009/0118488, and Drmanac et al., 2010, Science 327 (5961): 78-81. Briefly, after ligation of adapters to fragment the genomic library DNA to fragments, the ligated adapter fragments are circularized by ligation with a cyclic ligase and amplified in a rolling ring manner (as described by Lizardi et al., 1998. Nat. Genet. 19:225- 232 and in US 2007/0099208 Al). The elongated concatemeric structure of the amplicons promotes helicalization and thereby creates compact DNA nanospheres. DNA nanospheres can be captured on carriers, preferably to create an ordered or structured array such that distance between nanospheres is maintained to allow sequencing of individual DNA nanospheres. In some embodiments, such as those using the Complete Genomics (Mountain View, CA) method, successive rounds of adapter ligation, amplification, and digestion are performed before circularization to produce head-to-tail constructs having multiple genomic DNA fragments separated by adapter sequences.

[00190] Representative isothermal amplification techniques that can be used in the method of the disclosure include, but are not limited to, multi-substitution amplification (MDA) as described, for example, by Dean et al., Proc. Natl. Acad. sci. USA 99: 5261-66 (2002), or isothermal amplification of nucleic acids with strand exclusion, such as those described in US Patent No. 6,214,587. Other methods that are not based on PCR and can be used in the method of the disclosure of the invention are, for example, strand displacement amplification (SDA) as described, for example, by Walker et al., Molecular Methods for Detection Detection, Academic Press, Inc. ., 1995; US Pat. Nos. 5,455,166 and 5,130,238 and Walker et al., Nucl. Acids Res. 20: 1691-96 (1992), or hyperbranched exclusion amplification as described, for example, by Lage et al., Genome Res. 13: 294-307 (2003). Isothermal amplification methods can be carried out, for example, using Phi 29 polymerase for strand displacement, or using a large fragment of Bst DNA polymerase, 5'→3'-exo for randomized amplification of genomic DNA using primers. The advantage of using these polymerases is their high efficiency and strand displacement activity. High efficiency allows polymerases to form fragments of 10-20 kb in length. As mentioned above, smaller fragments can be obtained under isothermal conditions using polymerases having low chain displacement efficiency and activity, such as Klenow polymerase. Additional more detailed description of the amplification reactions and their conditions and components is given in US patent No. 7670810.

[00191] Another method for amplifying polynucleotides that is used in the present invention is PCR labeling, which uses a population of two-domain primers having a constant 5' region followed by a randomized 3' region, as described, for example, Grothues et al. Nucleic Acids Res. 21(5):1321-2 (1993). The first rounds of amplification are carried out to create a plurality of initiation sites on thermally denatured DNA based on individual hybridization based on a randomly synthesized 3' region. Based on the nature of the 3' region, it has been suggested that initiation sites are randomly distributed throughout the genome. Thereafter, unbound primers can be removed and further replication can be done using primers complementary to the 5' constant region.

[00192] In some embodiments, isothermal amplification can be performed by kinetic exclusion amplification (KEA), also referred to as exclusion amplification (ExAmp). A nucleic acid library of the invention can be obtained by a method comprising the step of reacting an amplification reagent to produce a plurality of amplification sites, each of which comprises a substantially clonal population of amplicons from a single target nucleic acid that occupies that site. In some embodiments of the invention, the amplification reaction is continued until a sufficient number of amplicons are generated to fill the corresponding amplification site. Filling an already occupied site to this level prevents anchorage and amplification of target nucleic acids at that site, leading to the formation of a clonal population of amplicons at that site. In some embodiments, apparent clonality can be achieved even if the amplification site is not filled to its full capacity before the second nucleic acid is anchored at that site. Under certain conditions, amplification of the first target nucleic acid may continue until a sufficient copy number is made to greatly out-compete or suppress copy-making of the second target nucleic acid that is being transported to that site. Thus, for example, in an embodiment of the invention that uses a bridged amplification method on a ring element with a diameter of less than 500 nm, it was determined that after 14 cycles of exponential amplification of the first target nucleic acid, contamination by the second target nucleic acid at the same site will produce an amount of contaminating amplicons that is insufficient to adversely affect sequencing analysis via synthesis on the Illumina sequencing platform.

[00193] In some embodiments of the invention, amplification sites in an array may be, but need not be, fully clonal. Rather, for some applications, a particular amplification site may be predominantly filled with amplicons from the first indexed fragment and may also have a low level of contaminating amplicons from the second target nucleic acid. The array may include one or more amplification sites that have a low level of contaminating amplicons, provided that the level of contamination will not adversely affect subsequent use of the array. So, for example, if the array is used for detection, then an acceptable level of contamination will be a level that will not adversely affect the signal-to-noise ratio or the resolution of the detection method. Accordingly, apparent clonality will, as usual, be related to the particular use or application of the array obtained by the methods described here. Typical levels of contamination that may be acceptable at a particular amplification site for specific applications include, but are not limited to, a maximum of 0.1%, 0.5%, 1%, 5%, 10%, or 25% of contaminating amplicons. The array may include one or more amplification sites having these representative levels of contaminating amplicons. For example, up to 5%, 10%, 25%, 50%, 75%, or even 100% of amplification sites in an array may have multiple contaminating amplicons. It is obvious that in an array or other collection of sites, at least 50%, 75%, 80%, 85%, 90%, 95%, or 99% or more of the sites may be clonal or may have apparent clonicity.

[00194] In some embodiments of the invention, the kinetic exclusion may occur when the process occurs at a sufficiently high rate and allows you to effectively exclude another event or another process. For example, one can create an array of nucleic acids in which the sites of the array are randomly filled with indexed fragments from solution, and copies of the indexed fragments are formed during amplification so that each of the fragments completely fills these sites. In accordance with the methods of kinetic exclusion according to the invention, the processes of filling and amplification can occur simultaneously under conditions in which the rate of amplification exceeds the rate of filling. Thus, for example, the relatively high rate at which copies are made at the site filled with the first target nucleic acid will effectively preclude the filling of the amplification site with the second nucleic acid. Kinetic exclusion amplification methods can be carried out in the manner detailed in US Patent Application Publication No. 2013/0338042.

[00195] Kinetic exclusion can use a relatively low rate of amplification initiation (eg, a slow rate of creation of the first copy of the indexed fragment) compared to a relatively high rate of creation of subsequent copies of the indexed fragment (or the first copy of the indexed fragment). In the example described in the previous paragraph, the kinetic exclusion is due to the relatively low rate of population of the indexed fragment (eg, relatively slow diffusion or transport) compared to the relatively high rate at which amplification occurs to fill the site with copies of the indexed fragment. In another illustrative embodiment of the invention, the kinetic exclusion may be due to slower formation of the first copy of the indexed fragment that fills the site (eg, delayed or slow activation) compared to the relatively high rate of filling this site with subsequent copies. In this example, a single site may be populated with several different indexed fragments (eg, several indexed fragments may be present in each site prior to amplification). However, the formation of the first copy of any given indexed fragment can be activated randomly, with the result that the average rate of formation of the first copy is relatively low compared to the rate at which subsequent copies are generated. In this case, although a single site may be filled with several different indexed fragments, however, kinetic exclusion allows only one of these indexed fragments to be amplified. More specifically, once the first indexed fragment is activated for amplification, the site will quickly fill up with copies to full capacity, which will prevent the creation of copies of the second indexed fragment at that site.

[00196] In one embodiment of the invention, a method is carried out for simultaneously (i) transporting indexed fragments to amplification sites at an average transport rate, and (ii) amplifying indexed fragments that are present at amplification sites at an average amplification rate, where the average amplification rate exceeds the average speed transport (US patent No. 9166913). Accordingly, in such embodiments of the invention, kinetic exclusion can be achieved using a relatively low transport rate. For example, to achieve the desired average transport rate, a sufficiently low concentration of indexable fragments can be chosen, and the lower the concentration, the lower the average transport rate. Alternatively or additionally, a high viscosity solution may be used to reduce the transport rate and/or molecular thickening agents may be present in the solution. Examples of suitable molecular thickening agents include, but are not limited to, polyethylene glycol (PEG), ficoll, dextran, or polyvinyl alcohol. Representative reagents and preparations for molecular thickening are described in US patent No. 7399590, which is introduced into the present description by reference. Another factor that can be adjusted to achieve the desired transport rate is the average size of the target nucleic acids.

[00197] The amplification reagent may include additional components that facilitate the formation of the amplicon, and in some cases, increase the rate of formation of the amplicon. An example is recombinase. The recombinase can facilitate amplicon formation through repeated entry/extension. More specifically, the recombinase can facilitate penetration of the indexed fragment by the polymerase, and primer extension by the polymerase with the indexed fragment used as a template to form the amplicon. This process can be repeated as a chain reaction, where the amplicons generated after each round of penetration/extension serve as templates in the next round. This process can be faster than standard PCR because it does not require a denaturation cycle (eg, heat or chemical denaturation). Thus, amplification initiated by recombinase can be carried out under isothermal conditions. It is generally desirable to include ATP or other nucleotides (or, in some cases, non-hydrolysable analogs) in the recombinase-stimulated amplification reagent to facilitate amplification. A mixture of recombinase and single chain binding protein (SSB) is particularly suitable because SSB can also stimulate amplification. Typical formulations for recombinase-facilitated amplification are those that are commercially available as TwistAmp kits from TwistDx (Cambridge, UK). Suitable reagent components for recombinase-facilitated amplification and reaction conditions are described in US Pat. Nos. 5,223,414 and 7,399,590.

[00198] Another example of a component that can be included in an amplification reagent to facilitate amplicon formation, and in some cases increase the rate of amplicon formation, is helicase. The helicase can facilitate amplicon formation by a chain reaction of amplicon formation. This process can be faster than standard PCR because it does not require a denaturation cycle (eg, heat or chemical denaturation). Thus, amplification initiated by helicase can be carried out under isothermal conditions. A mixture of helicase and single chain binding protein (SSB) is particularly suitable because SSB can also stimulate amplification. Exemplary preparations for helicase-facilitated amplification are commercially available preparations that are available as IsoAmp kits from Biohelix (Beverly, MA). In addition, examples of suitable formulations that include the helicase protein are described in US Pat. Nos. 7,399,590 and 7,829,284.

[00199] Another example of a component that can be included in an amplifying reagent to facilitate amplicon formation, and in some cases, to increase the rate of amplicon formation, is an origin binding protein.

Use in sequencing/Sequencing methods

[00200] After attaching the indexed fragments to the surface, the sequence of the immobilized and amplified indexed fragments is determined. Sequencing can be performed by any suitable sequencing method, and methods for determining the sequence of immobilized and amplified index fragments, including strand resynthesis, are known to those skilled in the art and are described, for example, by Bignell et al. (U.S. Patent 8, 053, 192), Gunderson et al. (WO 2016/130704), Shen et al. (US patent 8895249) and Pipenburg et al. (U.S. Patent 9309502).

[00201] The methods described herein can be used in combination with various nucleic acid sequencing methods. Particularly suitable methods are methods where nucleic acids are attached at fixed positions in an array such that their relative positions do not change, and methods where multiple array images are obtained. Particularly suitable are variants in which images are obtained in different color channels, for example, matching different labels used to distinguish one type of nucleotide base from another. In some embodiments of the invention, the method for determining the nucleotide sequence of an indexed fragment may be automated. Preferred embodiments of the invention include sequencing by synthesis ("SBS") methods.

[00202] SBS methods typically involve enzymatic extension of a growing nucleic acid strand by iterative addition of nucleotides to a template strand. In traditional SBS methods, a single nucleotide monomer can be presented to the target nucleotide in the presence of a polymerase with each delivery. However, in the methods described herein, more than one type of nucleotide monomer may be presented to the target nucleic acid in the presence of the polymerase during delivery.

[00203] In one embodiment of the invention, the nucleotide monomer includes blocked nucleic acids (LNA) or bridged nucleic acids (BNA). The use of LNA or BNA in the nucleotide monomer increases the efficiency of hybridization between the nucleotide monomer and the primer sequencing sequence present on the immobilized index fragment.

[00204] Nucleotide monomers that have a termination group or nucleotide monomers that do not contain a termination group can be used in SBS. Methods using nucleotide monomers that lack terminators include, for example, pyrosequencing and sequencing using γ-phosphate labeled nucleotides, as detailed hereinafter. In methods using nucleotide monomers that do not contain terminators, the number of nucleotides added in each cycle is usually variable and depends on the sequence of the template and the method of delivery of the nucleotides. For SBS methods that use nucleotide monomers having a terminating group, the terminator may be virtually irreversible under sequencing conditions, as is the case with traditional Sanger sequencing, which uses dideoxynucleotides, or the terminator may be reversible, as is the case in sequencing methods developed by the company Solexa (now called Illumina, Inc.).

[00205] SBS methods can use nucleotide monomers that are labeled or nucleotide monomers that are not labeled. Accordingly, turn-on events can be detected based on properties of the label, such as the fluorescence of the label; characteristic features of the nucleotide monomer, such as molecular weight or charge; a nucleotide incorporation by-product, such as releasable pyrophosphate or the like. In those embodiments where two or more different nucleotides are present in the sequencing reagent, the different nucleotides may be different from each other, or alternatively, the two or more different labels may be indistinguishable in the detection methods used. Thus, for example, different nucleotides present in a sequencing reagent may be labeled differently and can be distinguished using suitable optical instruments as described in the sequencing methods developed by Solexa (now called Illumina, Inc.).

[00206] Preferred embodiments of the invention include pyrosequencing techniques. Pyrosequencing detects the release of inorganic pyrophosphate (PPi) as specific nucleotides are incorporated into the growing chain (Ronaghi, M., Karamohamed, S., Pettersson, B., Uhlen, M. and Nyren, P. (1996) "Real-time DNA sequencing using detection of pyrophosphate release" Analytical Biochemistry 242(1), 84-9; Ronaghi, M. (2001) "Pyrosequencing sheds light on DNA sequencing" Genome Res. 11(1), 3-11; Ronaghi, M., Uhlen, M. and Nyren, P. (1998) "A sequencing method based on real-time pyrophosphate" Science 281(5375), 363; U.S. Patent Nos. 6210891; 6258568 and 6274320). In pyrosequencing, released PPi can be detected by direct conversion to adenosine triphosphate (ATP) by ATP sulphurase, and the level of ATP generated is determined by photons produced by luciferase. The nucleic acids to be sequenced can be attached to elements in the array, and the array can be visualized to capture chemiluminescent signals that are generated by incorporating the nucleotides into the elements of the array. An image can be obtained after processing the array with a specific type of nucleotide (eg, A, T, C, or G). The images obtained after the addition of each type of nucleotide will differ depending on the detected elements in the array. These differences in the images reflect the different contents of the element sequences in the array. However, the relative positions of each element remain unchanged in the images. Images can be stored, processed and analyzed using the methods described here. For example, images obtained after processing the array with each of the different types of nucleotides can be processed in the same manner as described here for images obtained from different detection channels in reversible terminator sequencing methods.

[00207] In another representative SBS, cycle sequencing is performed by stepwise addition of reversible terminator nucleotides containing, for example, a cleavable label or a photobleaching dye label, as described, for example, in WO 04/018497 and US Pat. No. 7057026. This method has been commercialized by Solexa (now called Illumina Inc.) and is also described in WO 91/06678 and WO 07/123,744. The availability of fluorescently labeled terminators, in which both ends can be reversible and the fluorescent label is cleaved, facilitates efficient sequencing via reversible cyclic termination (CRT). Polymerases can also be designed to efficiently incorporate modified nucleotides and extend the chain from these modified nucleotides.

[00208] In some variants of reversible terminator sequencing, labels do not substantially inhibit elongation under SBS reaction conditions. However, the detection marks can be removed, for example, by cleavage or degradation. Images can be made after incorporating labels into nucleic acid matrix elements. In specific embodiments of the invention, each cycle includes the simultaneous delivery of four different nucleotides in the array, and each type of nucleotide has a spectrally different label. Four images can then be obtained, each of which is obtained using a detection channel specific to one of four different labels. Alternatively, different types of nucleotides can be added sequentially and an image of the array can be obtained between each addition step. In such embodiments, each image will show nucleic acid elements that include nucleotides of a particular type. Different images will have or lack different elements due to the different composition of the sequence of each element. However, the relative position of the elements in the images remains unchanged. Images obtained using such SBS methods using a reversible terminator can be stored, processed and analyzed as described in this application. After the imaging step, labels and reversible terminator molecules can be removed for subsequent rounds of nucleotide addition and detection. Removing marks after they have been detected in a particular cycle and before the subsequent cycle may be advantageous in terms of reducing background signal and crosstalk between cycles. Examples of suitable labels and removal methods are provided in this application.

[00209] In specific embodiments of the invention, some or all of the nucleotide monomers may include reversible terminators. In such embodiments, reversible terminators/cleavable fluorophores may include fluorophores linked to the ribose molecule via a 3' ester bond (Metzker, Genome Res. 15: 1767-1776 (2005)). In other methods, the terminator chemistry method differs from the fluorescent label cleavage method (Ruparel et al., Proc Natl Acad Sci USA 102: 5932-7 (2005)). Ruparel et al. have described the preparation of reversible terminators that use a small 3'-allyl group to block elongation but can be easily deblocked by a short treatment with a palladium catalyst. The fluorophore was attached to the base via a photo-cleavable linker, which is readily cleaved by 30 seconds long wavelength UV light. Thus, a disulfide reduction agent or a photocleavage agent can be used as the cleavable linker. Another method of reversible termination is the use of natural termination, which will occur after bulk dye is placed on the dNTP. The presence of a charged bulk dye on dNTP can act as an effective terminator due to steric and/or electrostatic hindrance. The presence of one turn-on event prevents further turn-ons unless the dye is removed. Cleavage of the dye removes the fluorophore and effectively abolishes termination. Examples of modified nucleotides are also described in US Pat. Nos. 7427673 and 7057026.

[00210] Additional representative SBS systems and methods that may be used in conjunction with the methods and systems described herein are described in US Patent Application Publication Nos. 2007/0166705; 2006/0188901; 2006/0240439; 2006/0281109; PCT WO 06/064199 and WO 07/010251.

[00211] In some embodiments, the detection of four different nucleotides using less than four different labels can be used. For example, SBS can be performed using the methods and systems described in the included materials of US Patent Application Publication 2013/0079232. In the first example, a nucleotide pair can be detected at the same wavelength, but with only one difference, which is the difference in intensity for one member of the pair from the intensity for the other member of the pair, or in the replacement of one member of the pair (for example, by chemical modification, photochemical modification or physical modification), which causes the appearance or disappearance of the apparent signal in comparison with the signal detected for the other member of the pair. In another example, three out of four different nucleotides can be detected under certain conditions, while a fourth type nucleotide does not contain a label that can be detected under these conditions, or is detected at a minimum level under these conditions (for example, is detected at a minimum level due to background fluorescence, etc.). The inclusion of the first three nucleotides in the nucleic acid can be determined by the presence of the corresponding signals, and the inclusion of the fourth type of nucleotide in the nucleic acid can be determined by the absence or detection of any signal at a minimal level. In a third example, one type of nucleotide may include label(s) that(s) are(are) detected in two different channels, while other types of nucleotides are detected in no more than one of the channels. The above three representative configurations are not considered mutually exclusive and may be used in various combinations. A representative variant that combines all three examples is a fluorescence-based SBS method that uses a first type of nucleotide detectable in the first channel (for example, dATP having a label that is detected in the first channel when excited at the first excitation wavelength), a nucleotide of the second type, which is detected in the second channel (for example, dCTP, having a label that is detected in the second channel when excited at a second excitation wavelength), a nucleotide of the third type, which is detected in the first and second channel (for example, dTTP, having at least one label that is detected in both channels when excited at the first and/or second excitation wavelength), and a fourth type nucleotide that does not contain a label, that is, is not detected or is detected at a minimum level in any channel (for example, dGTP, unlabeled).

[00212] Further, as described in the included materials of US Patent Application Publication No. 2013/0079232, sequencing data can be obtained using a single channel. In such so-called single-dye sequencing methods, the first type nucleotide is labeled, but the label is removed after the first image is acquired, and the second type nucleotide is labeled only after the first image is obtained. The third type nucleotide retains its label in both the first and second images, while the fourth type nucleotide remains unlabeled in both images.

[00213] In some embodiments of the invention, sequencing by ligation methods can be performed. Such methods use DNA ligase to incorporate oligonucleotides and identify the inclusion of such oligonucleotides. Oligonucleotides typically have different labels that correlate with the identity of the particular nucleotide in the sequence to which the oligonucleotides hybridize. As with other SBS methods, images can be obtained after processing an array of nucleic acid elements with labeled sequencing reagents. Each image will show nucleic acid elements having a particular type of label included. Different elements will be present or absent in different images due to the different sequence content of each element, but the relative position of the elements in these images will remain the same. Images obtained by ligation-based sequencing methods can be stored, processed and analyzed as described in this application. Representative SBS systems and methods that can be used in conjunction with the methods and systems described herein are described in US Pat.

[00214] In some embodiments, nanopore sequencing can be performed (Deamer, D. W. & Akeson, M. "Nanopores and nucleic acids: prospects for ultrarapid sequencing." Trends Biotechnol. 18, 147-151 (2000); Deamer, D. and D. Branton, "Characterization of nucleic acids by nanopore analysis", Acc. Chem. Res. 35:817-825 (2002), Li, T, M. Gershow, D. Stein, E. Brandin, and J. A. Golovchenko, "DNA molecules and configurations in a solid-state nanopore microscope" Nat. Mater. 2:611-615 (2003)). In such embodiments of the invention, the indexed fragment passes through the nanopore. The nanopore can be a synthetic pore or a biological membrane protein such as α-hemolysin. If the fragment being indexed passes through a nanopore, then each base pair can be identified by measuring the fluctuations in the electrical conductivity of the pore (U.S. Patent No. 7001792; Soni, G. V. & Meller, "A. Progress towards ultrafast DNA sequencing using solid-state nanopores." Clin. Chem 53, 1996-2001 (2007), Healy, K. "Nanopore-based single-molecule DNA analysis", Nanomed 2, 459-481 (2007), Cockroft, S. L., Chu, J., Amorin, M. & Ghadiri, M. R. "A single-molecule nanopore device detects DNA polymerase activity with single-nucleotide resolution", J. Am. Chem. Soc. 130, 818-820 (2008)). Data obtained after nanopore sequencing can be stored, processed and analyzed as described in this application. In particular, the data may be processed as an image in accordance with the representative processing of optical images and other images described in this application.

[00215] In some embodiments of the invention, methods can be used that include real-time monitoring of DNA polymerase activity. Nucleotide inclusions can be detected by fluorescent resonance energy transfer (FRET) interactions between a fluorophore-containing polymerase and γ-phosphate-labeled nucleotides, as described, for example, in US Pat. 7,329,492 and 7,211,414 or nucleotide inclusions can be detected using zero-mode waveguides as described, for example, in US Pat. 7,315,019 and using fluorescent nucleotide analogs and engineered polymerases as described, for example, in US Pat. 7405281 and US Patent Application Publication No. 2008/0108082. Irradiation can be limited to a heptoliter-scale volume surrounded by surface-bound polymerase, so that incorporation of fluorescently labeled nucleotides can be observed at a low background level (Levene, M. J. et al. "Zero-mode waveguides for single molecule analysis at high concentrations ", Science 299, 682-686 (2003); Lundquist, P. M. et al. "Parallel confocal detection of single molecules in real time", Opt. Lett. 33, 1026-1028 (2008); Korlach, J. et al. , "Selective aluminum passivation for targeted immobilization of single DNA polymerase molecules in zero-mode waveguide nano structures)", Proc. Natl. Acad. sci. USA 105, 1176-1181 (2008)). Images obtained by such methods can be stored, processed and analyzed as described in this application.

[00216] Some variants of SBS include detection of a proton released after incorporation of a nucleotide into the extension product. For example, when sequencing based on the detection of released protons, an electrical detector and methods using it can be used, which are commercially available and developed by Ion Torrent (Guilford, CT, a Life Technologies subsidiary) or sequencing methods and systems described in application publications. for a US patent. No. 2009/0026082; 2009/0127589; 2010/0137143; and 2010/0282617. The methods described here for amplifying target nucleic acids by kinetic exclusion can be readily applied to substrates used for proton detection. More specifically, the methods described herein can be used to generate clonal populations of amplicons that are used to detect protons.

[00217] The above SBS methods can advantageously be implemented in a multiplex format such that many different indexed fragments are modified at the same time. In specific embodiments of the invention, different indexed fragments can be processed in a common reaction vessel or on the surface of a specific base. This facilitates the delivery of sequencing reagents, the removal of unreacted reagents, and the detection of turn-on events in a multiplexed manner. In embodiments of the invention that use surface-bound target nucleic acids, the indexed fragments may be in array format. In an array format, indexed fragments can typically be associated with a surface in a wide variety of distinguishable ways. The indexed moieties may be linked via a direct covalent bond, by attachment to a sphere or other particle, or by binding to a polymerase or other molecule that is bound to a surface. The array may include one copy of the indexed fragment per site (also called element), or multiple copies of the same sequence may be present in each site or element. Multiple copies can be obtained by amplification methods such as bridge amplification or emulsion PCR, as described in more detail in this application.

[00218] The methods described herein may use arrays having elements of various densities, including, for example, at least about 10 elements/cm ² , 100 elements/cm ² , 500 elements/cm ² , 1000 elements/cm ² , 5,000 cells/cm ² , 10,000 cells/cm ² , 50,000 cells/cm ² , 100,000 cells/cm ² , 1,000,000 cells/cm ² , 5,000,000 cells/cm ² or more.

[00219] The advantage of the methods described here is that they simultaneously provide fast and efficient detection of multiple elements per cm ² . Accordingly, the present invention relates to integrated systems that allow the production and detection of nucleic acids by methods known to those skilled in the art, such as the methods described in this application. Thus, an integrated system according to the invention may include liquid components capable of delivering amplification reagents and/or sequencing reagents to one or more immobilized indexed fragments, where said system includes components such as pumps, valves, reservoirs, flow lines, and the like. . The flow cell may be configured and/or used in an integrated system for the detection of target nucleic acids. Representative flow cells are described, for example, in US Patent Application Publication Nos. 2010/0111768 and 13/273666. As illustrated for the flow cells, one or more liquid components of the integrated system can be used in the amplification and detection method. An exemplary embodiment of nucleic acid sequencing is where one or more liquid components of the integrated system can be used in the amplification method described herein and to deliver sequencing reagents in the sequencing method, such as the method illustrated above. Alternatively, the integrated system may include separate fluid systems for performing amplification methods and detection methods. Examples of integrated sequencing systems suitable for obtaining amplified nucleic acids as well as nucleic acid sequencing include, but are not limited to, the MiSeqTM platform (Illumina, Inc., San Diego, CA) and the devices described in US Pat. no. 13/273666.

[00220] The present invention also relates to compositions. With the practical application of the methods described here, various compositions can be obtained. For example, a composition can be prepared comprising indexed nucleic acid fragments, wherein the indexed nucleic acid fragments are derived from newly synthesized RNA. In one embodiment of the invention, the newly synthesized RNA is labeled. The present invention also relates to a multiwell plate, wherein the well of the multiwell plate contains indexed nucleic acid fragments.

[00221] The present invention also relates to kits. In one embodiment of the invention, the kit is designed to obtain a sequencing library, where the newly synthesized RNA is labeled. In another embodiment of the invention, the kit includes a nucleotide tag as described herein. In another embodiment of the invention, the kit includes one or more RNA annealing primers, wherein at least one primer is designed to target amplify one or more predefined nucleic acids. In a further embodiment of the invention, the kit includes components for adding at least three indices to nucleic acids. The kit may also include other components suitable for obtaining a sequencing library. For example, the kit may include at least one enzyme that mediates ligation, primer extension, or amplification to process RNA molecules to include an index. The kit may include nucleic acids with index sequences. The kit may also include other components suitable for adding an index to the nucleic acid, such as a transposome complex. The kit may also include one or more RNA annealing primers. Primers may be designed to generate the entire transcriptome (eg, a primer that includes a poly-T sequence) or to target amplification of one or more predefined nucleic acids.

[00222] The components of the kit in a suitable packaging material are usually present in an amount sufficient to conduct at least one analysis or its application. In this case, may be included, but not necessarily, and other components, such as buffers and solutions. Typically, instructions for use of the packaged components are also included. As used herein, the term "packaging material" means one or more physical structures used to contain the contents of a kit. The packaging material is made by routine methods, usually to provide a sterile environment without impurities. The packaging material may have a label stating that the components can be used to make a sequencing library. In addition, the packaging material contains instructions for the use of the materials contained in the kit. As used herein, the term "packaging" means a container of materials such as glass, plastic, paper, foil, etc., capable of keeping the components of a kit within fixed limits. "Instructions for use" usually includes a well-known expression describing the concentration of a reagent or at least one of the parameters of the analytical method, such as the relative amounts of reagent and sample to be mixed, storage time of reagent/sample mixtures, temperature, buffer conditions, and the like.

Representative Embodiments of the Invention

[00223] Option 1. A method of obtaining a library for sequencing containing nucleic acids derived from a plurality of individual nuclei or cells, where this method includes:

(a) obtaining a plurality of nuclei or cells in a first plurality of compartments,

where each compartment contains a subpopulation of nuclei or cells;

(b) labeling newly synthesized RNA in subpopulations of cells or nuclei derived from cells;

(c) processing the RNA molecules in each subset of nuclei or cells to obtain indexed nuclei or cells,

where the processing includes adding to the RNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed DNA nucleic acids present in the indexed nuclei or cells,

where the processing includes ligation, primer extension, hybridization or amplification; and

(d) combining indexed nuclei or cells to form combined indexed nuclei or cells.

[00224] Option 2. The method of Option 1, where the processing includes:

contacting subpopulations with reverse transcriptase and a primer that hybridize with RNA nucleic acids to form double-stranded DNA nucleic acids containing the primer and the corresponding DNA nucleotide sequence from RNA template molecules.

[00225] Option 3. The method of options 1 or 2, wherein the primer contains a poly-T nucleotide sequence that hybridizes to the mRNA poly(A) tail.

[00226] Option 4. The method of any of options 1-3, wherein the treatment also includes contacting the subpopulations with a second primer, where the second primer contains a sequence that hybridizes to a predetermined DNA nucleic acid.

[00227] Option 5. The method of any of options 1-4, where the second primer contains a compartment-specific index.

[00228] Option 6. The method of any of options 1-5, wherein the primer contains a sequence that hybridizes to a predetermined RNA nucleic acid.

[00229] Option 7. The method of any of options 1-6, wherein the method includes primers in different compartments that hybridize to different nucleotides of the same predefined RNA nucleic acid.

[00230] Option 8. The method of any of options 1-7, wherein the primer includes a template switch primer.

[00231] Option 9. The method of any of options 1-7, wherein the processing to add the first compartment-specific index sequence comprises a two-step method of adding a nucleotide sequence containing

a universal sequence to RNA nucleic acids to obtain DNA nucleic acids, and then adding a first compartment-specific index sequence to DNA nucleic acids.

[00232] Option 10. A method of obtaining a sequencing library containing nucleic acids derived from a plurality of individual nuclei or cells, wherein said method comprises:

(a) obtaining a plurality of nuclei or cells in a first plurality of compartments,

where each compartment contains a subpopulation of nuclei or cells;

(b) contacting each subpopulation with a reverse transcriptase and a primer that hybridizes to a predetermined RNA nucleic acid to form double-stranded DNA nucleic acids containing the primer and the corresponding DNA nucleotide sequence from the RNA template molecules;

(c) processing DNA molecules in each subset of nuclei or cells to obtain indexed nuclei or cells,

where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed DNA nucleic acids,

present in indexed nuclei or cells,

where the processing includes ligation, primer extension, hybridization or amplification; and

(d) combining indexed nuclei or cells to form combined indexed nuclei or cells.

[00233] Option 11. The method of option 10, where the primer contains the first compartment-specific index sequence.

[00234] Option 12. The method of options 10 or 11, further comprising, before contacting, labeling newly synthesized RNA in subpopulations of cells or nuclei isolated from cells.

[00235] Option 13. The method of any of options 10-12, wherein the processing for adding the first compartment-specific index sequence comprises a two-step method of adding a nucleotide sequence containing a universal sequence to nucleic acids, and then adding the first compartment-specific index sequence to nucleic acids. acids.

[00236] Option 14. The method of any of options 1-13, wherein the predetermined RNA nucleic acid is mRNA.

[00237] Option 15. The method of any of options 1-14, wherein pre-existing RNA nucleic acids and newly synthesized RNA nucleic acids are labeled with the same indices in the same compartment.

[00238] Option 16. The method of any of options 1-15, where the labeling includes incubating a plurality of nuclei or cells in a composition containing a nucleotide tag, where the nucleotide tag is included in the newly synthesized RNA.

[00239] Option 17. The method of any of options 1-16, wherein the nucleotide tag comprises a nucleotide analog, a hapten-labeled nucleotide, a mutagenic nucleotide, or a nucleotide that can be modified by a chemical reaction.

[00240] Option 18. The method of any of options 1-17, wherein more than one nucleotide tag is included in the newly synthesized RNA.

[00241] Option 19. The method of any of options 1-18, wherein the ratios of the nucleotide tag or tags are different for different compartments or time points.

[00242] Option 20. The method of any of options 1-19, also comprising treating subpopulations of nuclei or cells under predetermined conditions prior to labeling.

[00243] Option 21. The method of any of options 1-20, where the predetermined condition includes processing by an agent.

[00244] Option 22. The method of any of options 1-21, wherein the agent comprises a protein, a non-ribosomal protein, a polyketide, an organic molecule, an inorganic molecule, an RNA or RNAi molecule, a carbohydrate, a glycoprotein, a nucleic acid, or a combination thereof.

[00245] Option 23. The method of any of options 1-22, wherein the agent comprises a therapeutic drug.

[00246] Option 24. The method of any of options 1-23, where the predetermined conditions of the two or more compartments are different.

[00247] Option 25. The method of any of options 1-24, where the treatment and labeling are carried out simultaneously or the treatment is carried out before labeling.

[00248] Option 26. The method of any of options 1-25, further comprising:

distributing the subpopulations of the combined indexed nuclei or cells over the second set of compartments and adding to the indexed nucleic acids present in the subpopulations of nuclei or cells a second index sequence to obtain nuclei or cells with two indexes containing nucleic acid fragments with two indexes, where the addition includes ligation, primer extension, hybridization, amplification or transposition; and

combining nuclei or cells with two indices to obtain combined nuclei or cells with two indices.

[00249] Option 27. The method of any of options 1-26, further comprising:

distributing the subpopulations of fused nuclei or cells with two indexes over a third set of compartments and adding to the indexed nucleic acids present in the subpopulations of nuclei or cells with a third index sequence to obtain nuclei or cells with three indexes containing nucleic acid fragments with three indexes, where the addition includes ligation, primer extension, hybridization, amplification or transposition; and

combining nuclei or cells with three indices to obtain combined nuclei or cells with three indices.

[00250] Option 28. The method of any of options 1-27, where the distribution includes breeding.

[00251] Option 29. The method of any of options 1-27, where the distribution includes sorting.

[00252] Option 30. The method of any of options 1-29, wherein the addition comprises contacting the subpopulations with the hairpin ligation duplex under conditions suitable for ligating the hairpin ligation duplex to the end of nucleic acid fragments containing one or two index sequences.

[00253] Option 31. The method of any of options 1-30, wherein the addition comprises contacting nucleic acid fragments containing one or more index sequences with a transposome complex, where the transposome complex in compartments contains a transposase and a universal sequence, where contacting also includes the conditions, suitable for fragmenting nucleic acid fragments and incorporating a universal sequence into nucleic acid fragments.

[00254] Option 32. The method of any of options 1-31, wherein the addition comprises ligating a first compartment-specific index sequence, as well as adding a second index sequence to obtain nuclei or cells with two indexes containing nucleic acid fragments with two indexes, where adding includes transposition.

[00255] Option 33. The method of any of options 1-32, wherein the addition comprises ligating a second compartment-specific index sequence, as well as adding a third index sequence to produce nuclei or cells with two indexes containing nucleic acid fragments with three indexes, where adding includes transposition.

[00256] Option 34. The method of any of options 1-33, where the compartment includes a well or drop.

[00257] Option 35. The method of any of options 1-34, where the compartments of the first set of compartments contain from 50 to 100,000,000 nuclei or cells.

[00258] Option 36. The method of any of options 1-35, where the compartments of the second set of compartments contain from 50 to 100,000,000 nuclei or cells.

[00259] Option 37. The method of any of options 1-36, where the compartments of the third set of compartments contain from 50 to 100,000,000 nuclei or cells.

[00260] Option 38. The method of any of options 1-37, further comprising obtaining indexed nucleic acids from the pooled indexed nuclei or cells, obtaining a library for sequencing from a plurality of nuclei or cells.

[00261] Option 39. The method of any of options 1-38, further comprising obtaining dual-index nucleic acids from pooled nuclei or dual-index cells to obtain a sequencing library from a plurality of nuclei or cells.

[00262] Option 40. The method of any of options 1-39, further comprising generating tri-index nucleic acids from tri-index pooled nuclei or cells to obtain a sequencing library from a plurality of nuclei or cells.

[00263] Option 41. The method of any of options 1-40, further comprising:

obtaining a surface containing a plurality of amplification sites,

where the sites of amplification include at least two populations of linked single-stranded capture oligonucleotides having a free 3' end, and

contacting a surface containing amplification sites with nucleic acid fragments containing one, two or three index sequences under conditions suitable to obtain a plurality of amplification sites, each containing a clonal population of amplicons, from a single fragment containing a plurality of indices.

[00264] Option 42. The method of any of options 1-41, wherein adding the compartment-specific index sequence comprises a two-step method of adding a nucleotide sequence containing the universal sequence to the nucleic acids, and then adding the first compartment-specific index sequence to the nucleic acids.

[00265] Option 43. A method of obtaining a sequencing library containing nucleic acids derived from a plurality of individual nuclei or cells, wherein said method comprises:

(a) obtaining a plurality of nuclei or cells in a first plurality of compartments,

where each compartment contains a subpopulation of nuclei or cells;

(b) contacting each subpopulation with a reverse transcriptase and a primer to obtain double-stranded DNA nucleic acids containing the primer and the corresponding DNA nucleotide sequence from template RNA molecules;

(c) processing DNA molecules in each subset of nuclei or cells to obtain indexed nuclei or cells,

where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed nucleic acids present in the indexed nuclei or cells,

where the processing includes ligation, primer extension, hybridization, amplification or transposition;

(d) combining indexed nuclei or cells to form combined indexed nuclei or cells;

(e) distributing the pooled indexed nuclei or cells across a second set of compartments, where each compartment contains a subset of nuclei or cells;

(f) processing DNA molecules in each subset of nuclei or cells to obtain nuclei or cells with two indices,

where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, a second compartment-specific index sequence to obtain nucleic acids with two indices present in the indexed nuclei or cells,

where the processing includes ligation, primer extension, hybridization, amplification or transposition;

(g) combining nuclei or cells with two indices to obtain combined nuclei or cells with two indices;

(h) the distribution of the combined nuclei or cells with two indices across the third set of compartments,

where each compartment contains a subpopulation of nuclei or cells;

(i) processing DNA molecules in each subset of nuclei or cells to obtain nuclei or cells with three indices,

where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, a third compartment-specific index sequence to obtain nucleic acids with three indices present in the indexed nuclei or cells,

where the processing includes ligation, primer extension, hybridization, amplification or transposition; and

(j) combining nuclei or cells with three indices to obtain combined nuclei or cells with three indices.

[00266] Option 44. A method of obtaining a sequencing library containing nucleic acids derived from a plurality of individual nuclei or cells, wherein said method comprises:

(a) obtaining a plurality of nuclei or cells;

(b) contacting a plurality of nuclei or cells with a reverse transcriptase and a primer to produce double-stranded DNA nucleic acids containing the primer and the corresponding DNA nucleotide sequence from the RNA template nucleic acids;

(c) the distribution of nuclei or cells in the first set of compartments,

where each compartment contains a subpopulation of nuclei or cells;

(d) processing DNA molecules in each subset of nuclei or cells to obtain indexed nuclei or cells,

where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed nucleic acids present in the indexed nuclei or cells,

where the processing includes ligation, primer extension, hybridization, amplification or transposition;

(e) combining indexed nuclei or cells to form combined indexed nuclei or cells;

(f) distributing the pooled indexed nuclei or cells across a second set of compartments, where each compartment contains a subset of nuclei or cells;

(g) processing DNA molecules in each subset of nuclei or cells to obtain nuclei or cells with two indices,

where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, a second compartment-specific index sequence to obtain nucleic acids with two indices present in the indexed nuclei or cells,

where the processing includes ligation, primer extension, hybridization, amplification or transposition;

(h) combining nuclei or cells with two indices to obtain combined nuclei or cells with two indices;

(i) distributing pooled nuclei or cells with two indices across a third set of compartments, where each compartment contains a subset of nuclei or cells;

(j) processing DNA molecules in each subset of nuclei or cells to obtain nuclei or cells with three indices,

where the processing includes adding to the DNA nucleic acids present in subpopulations of nuclei or cells, a third compartment-specific index sequence to obtain nucleic acids with three indices present in the indexed nuclei or cells,

where the processing includes ligation, primer extension, hybridization, amplification or transposition; and

(k) merging nuclei or cells with three indices to obtain combined nuclei or cells with three indices.

[00267] Option 45. The method of any of options 43 or 44, wherein the primer hybridizes to RNA nucleic acids to form double-stranded DNA nucleic acids containing the primer and the corresponding DNA nucleotide sequence from RNA template molecules.

[00268] Option 46. The method of any of options 43-45, wherein the primer contains a poly-T nucleotide sequence that hybridizes to the poly(A) tail of the mRNA.

[00269] Option 47. The method of any of options 43-46, where contacting also includes contacting subpopulations with a second primer, where the second primer contains

a sequence that hybridizes to a predetermined DNA nucleic acid.

[00270] Option 48. The method of any of options 43-47, where the second primer contains a compartment-specific index.

[00271] Option 49. The method of any of options 43-45, wherein the primer contains a sequence that hybridizes to a predetermined RNA nucleic acid.

[00272] Option 50. The method of any of options 43-49, wherein the predetermined RNA nucleic acid is mRNA.

[00273] Option 51. The method of any of options 43-50, wherein the primer includes a template switch primer.

[00274] Option 52. The method of any of options 43-51, wherein the processing to add one or more first, second, or third compartment-specific index sequences comprises a two-step method of adding a nucleotide sequence containing the universal sequence to nucleic acids, and then adding the first compartment-specific index sequence for DNA nucleic acids.

[00275] Option 53. The method of any of options 43-52, wherein the primer contains a first compartment-specific index sequence.

[00276] Option 54. The method of any of options 43-53, further comprising, before contacting, labeling newly synthesized RNA in subpopulations of cells or nuclei isolated from cells.

[00277] Option 55. The method of any of options 43-54, wherein pre-existing RNA nucleic acids and newly synthesized RNA nucleic acids are labeled with the same index in the same compartment.

[00278] Option 56. The method of any of options 43-55, wherein the labeling comprises incubating a plurality of nuclei or cells in a composition containing a nucleotide tag, wherein the nucleotide tag is incorporated into the newly synthesized RNA.

[00279] Option 57. The method of any of options 43-56, wherein the nucleotide tag comprises a nucleotide analog, a hapten-labeled nucleotide, a mutagenic nucleotide, or a nucleotide that can be modified by a chemical reaction.

[00280] Option 58. The method of any of options 43-57, wherein more than one nucleotide tag is included in the newly synthesized RNA.

[00281] Option 59. The method of any of options 43-58, wherein the ratios of the nucleotide tag or tags are different for different compartments or time points.

[00282] Option 60. The method of any of options 43-59, also comprising treating subpopulations of nuclei or compartment cells under predetermined conditions prior to labeling.

[00283] Option 61. The method of any of options 43-60, where the predetermined condition includes processing by an agent.

[00284] Option 62. The method of any of options 43-61, wherein the agent comprises a protein, a non-ribosomal protein, a polyketide, an organic molecule, an inorganic molecule, an RNA or RNAi molecule, a carbohydrate, a glycoprotein, a nucleic acid, or a combination thereof.

[00285] Option 63. The method of any of options 43-62, wherein the agent comprises a therapeutic drug.

[00286] Option 64. The method of any of options 43-63, where the predetermined conditions of the two or more compartments are different.

[00287] Option 65. The method of any of options 43-64, where the treatment and labeling are carried out simultaneously or the treatment is carried out before labeling.

[00288] Option 66. The method of any of options 43-65, where one or more distributions include dilution.

[00289] Option 67. The method of any of options 43-65, where one or more distributions include sorting.

[00290] Option 68. The method of any of options 43-67, wherein adding one or more of the first, second, or third compartment-specific index sequences comprises contacting the subpopulations with the hairpin ligation duplex under conditions suitable for ligating the hairpin ligation duplex to the end of the nucleic acid fragments. acids.

[00291] Option 69. The method of any of options 43-68, wherein adding one or more of the first, second, or third compartment-specific index sequences comprises contacting nucleic acid fragments with a transposome complex, where the transposome complex in the compartments contains a transposase and a universal sequence, where contacting also includes conditions suitable for fragmenting nucleic acid fragments and incorporating the nucleotide sequence into nucleic acid fragments.

[00292] Option 70. The method of any of options 43-69, wherein adding the first or second compartment-specific index involves ligation and adding a subsequent compartment-specific index sequence involves transposition.

[00293] Option 71. The method of any of options 43-70, where the compartment includes a well or drop.

[00294] Option 72. The method of any of options 43-71, where the compartments of the first set of compartments contain from 50 to 100,000,000 nuclei or cells.

[00295] Option 73. The method of any of options 43-72, where the compartments of the second set of compartments contain from 50 to 100,000,000 nuclei or cells.

[00296] Option 74. The method of any of options 43-73, where the compartments of the third set of compartments contain from 50 to 100,000,000 nuclei or cells.

[00297] Option 75. The method of any of embodiments 43-74, further comprising obtaining tri-index nucleic acids from tri-index pooled nuclei or cells, and thereby obtaining a sequencing library from multiple nuclei or cells.

[00298] Option 76. The method of any of options 43-76, further comprising:

obtaining a surface containing a plurality of amplification sites,

where the sites of amplification include at least two populations of linked single-stranded capture oligonucleotides having a free 3' end, and

contacting a surface containing amplification sites with nucleic acid fragments with three indexes under conditions suitable for obtaining a plurality of amplification sites, each containing a clonal population of amplicons, from a single fragment containing a plurality of indices.

[00299] Option 77. A method of obtaining a sequencing library containing nucleic acids from a plurality of individual cells, where the method includes:

(a) obtaining nuclei from a plurality of cells;

(b) distributing nuclear subpopulations across a first plurality of compartments and contacting each subpopulation with a reverse transcriptase and a primer, where the primer in each compartment contains a first index sequence that is different from first index sequences in other compartments, to obtain indexed nuclei containing indexed nucleic acid fragments;

(c) combining indexed kernels to obtain combined indexed kernels;

(d) allocating subpopulations of the pooled indexed nuclei to a second plurality of compartments and contacting each subpopulation with the hairpin ligation duplex under conditions suitable for ligating the hairpin ligation duplex to the end of the indexed nucleic acid fragments containing the first index sequence to produce dual-index nuclei containing the fragments a nucleic acid with two indexes, where the hairpin ligation duplex contains a second index sequence different from the second index sequences in other compartments;

(e) merging the 2-index kernels to produce merged 2-index kernels;

(f) distributing the subpopulations of the dual-index pooled nuclei across a third set of compartments and processing the dual-index nucleic acid fragments under conditions suitable for second strand synthesis;

(g) contacting the two-index nucleic acid fragments with a transposome complex, wherein the transposome complex in each compartment contains a transposase and a universal sequence, where the contacting includes conditions suitable for fragmenting the two-index nucleic acid fragments and incorporating the universal sequence into the two-index nucleic acid fragments. indices to obtain nucleic acid fragments with two indices, containing the first and second indices at one end and the universal sequence at the other end;

(h) inserting into nucleic acid fragments with two indexes in each compartment a third index sequence to obtain fragments with three indexes;

(i) combining fragments with three indexes to obtain a sequencing library containing transcriptome nucleic acids from a plurality of individual cells.

[00300] Option 78. The method of option 77, wherein the primer contains a poly-T nucleotide sequence that hybridizes to the poly(A) tail of the mRNA.

[00301] Option 79. The method of options 77-78, where the primer of each compartment contains a sequence that hybridizes to a predetermined mRNA.

[00302] Option 80. The method of any of options 77-79, wherein the method includes primers in different compartments that hybridize to different nucleotides of the same predefined mRNA.

[00303] Option 81. A method of obtaining a transcriptome sequencing library containing nucleic acids from a plurality of single cells, wherein said method comprises:

(a) obtaining pooled nuclei from a plurality of cells;

(b) contacting the pooled nuclei with reverse transcriptase and a primer containing the oligo-dT sequence, which hybridizes to the mRNA poly(A) tail to form pooled nuclei containing nucleic acid fragments;

(c) distributing subpopulations of pooled nuclei across a plurality of compartments and contacting each subpopulation with a hairpin ligation duplex under conditions suitable for ligating the hairpin ligation duplex to the end of the nucleic acid fragments to produce indexed nuclei containing indexed nucleic acid fragments, where the hairpin ligation duplex contains an index sequence , different from index sequences in other compartments;

(d) combining indexed kernels to obtain combined indexed kernels;

(e) allocating subpopulations of the pooled indexed nuclei to a second set of compartments and processing the indexed nucleic acid fragments under conditions suitable for second strand synthesis;

(f) contacting the indexed nucleic acid fragments with a transposome complex, where the transposome complex in each compartment contains a transposase and a universal sequence, where the contacting includes conditions suitable for fragmenting the indexed nucleic acid fragments and incorporating the universal sequence into the indexed nucleic acid fragments to obtain indexed nucleic acid fragments acids containing an index at one end and a universal sequence at the other end;

(g) inserting into the indexed nucleic acid fragments in each compartment a second index sequence to produce fragments with two indexes;

(j) combining fragments with two indexes to obtain a sequencing library containing transcriptome nucleic acids from a plurality of individual cells.

[00304] Option 82. A method for isolating nuclei, including:

(a) flash freezing the tissue in liquid nitrogen;

(b) reducing the size of the tissue to obtain a treated tissue; and

(c) recovering nuclei from the treated tissue by incubation in a buffer that stimulates cell lysis and maintains the integrity of the nuclei in the absence of one or more exogenous enzymes.

[00305] Embodiment 83. The method of Embodiment 82, wherein said reduction in size comprises tissue shredding, exposing the tissue to physical force, or a combination thereof.

[00306] Option 84. The method of option 82 or 83, further comprising:

(d) treating the recovered nuclei with a cross-linking agent to obtain fixed nuclei; and

(e) washing fixed cores.

[00307] Option 85. A kit for its use in order to obtain a library for sequencing, where the specified kit includes a nucleotide tag and at least one enzyme that mediates ligation, primer extension or amplification.

[00308] Option 86. A kit for its use in order to obtain a sequencing library, where the specified kit includes a primer that hybridizes to a predetermined nucleic acid, and at least one enzyme that mediates ligation, primer extension or amplification.

Examples

[00309] The present invention is illustrated by the following examples. It should be noted that specific examples, materials, quantities and procedures should be interpreted in a broad sense in accordance with the scope and essence of the invention disclosed here.

Example 1

[00310] Dynamic transcription pattern in mammalian organogenesis with single cell resolution.

[00311] During mammalian organogenesis, the cells of the three germ layers transform into an embryo that includes most of the major internal and external organs. Key regulators of developmental defects can be explored at this important stage, but current technologies lack the throughput and resolution sufficient to provide general information about the molecular states and trajectories of a rapidly diversifying and expanding number of different cell types. The authors attempted to investigate the dynamics of transcription during the development of mice during organogenesis with resolution of individual cells. After improving the protocol based on combinatorial indexing of single cells (“sci-RNA-seq3”), the authors profiled over 2 million cells obtained from 61 mouse embryos at the gestation stage at days 9.5-13.5 (E9.5 -E13.5; 10-15 repetitions at each time point). The authors have identified hundreds of cells of expanding, contracting and transitional types, many of which are detected only by the depth of scanning coverage of cells defined here, and the corresponding sets of marker genes specific to cells of a particular type have been identified, some of which were confirmed by the authors by the general level of their hybridization in situ. The authors investigated the dynamics of proliferation and expression of genes in specific cell types over a certain period of time, including targeted analyzes of only the apical ectodermal ridge, limb mesenchyme, and skeletal muscle. Using the new algorithm, the authors identified the main developmental pathways of individual cells during murine organogenesis and found examples of different pathways to reach the same endpoint, i.e. branching and convergence. These data constitute a major resource in mammalian developmental biology and are presented in a way that makes it easier to understand the information that scientists are constantly commenting on.

[00312] Introduction

[00313] Mammalian organogenesis is an amazing process. Within a short period of time, the cells of the three germ layers develop into a mature embryo, which includes most of the major internal and external organs. Although human embryos at a very early stage can be cultured and examined in vitro ¹ , however, access to materials corresponding to the later stages of human embryonic development is limited. And therefore, most studies of mammalian organogenesis are carried out on model organisms, and, in particular, on mice.

[00314] Compared to humans, mice develop rapidly, with only 21 days between fertilization and birth. Implantation of mouse blastocysts (32-64 cells) occurs on the 4th day of embryogenesis (E4.0). This stage is followed by gastrulation and the formation of primary germ layers (E6.5-E7.5; 660-15K cells) ^2,3 . During this period of time, the formation of primitive bands and the distribution of individual lines of differentiation of the embryo from the anterior to the posterior region ⁴ occur. In the early somite stages (E8.0-E8.5), the embryo transitions from gastrulation to early organogenesis associated with the formation of the neural plate and cardiac tube (60K-90K cells). Classical organogenesis begins at the E9.5 stage. In the next four days (E9.5-E13.5), the mouse embryo develops from a few hundred thousand cells to more than ten million cells, and sensory organs, the gastrointestinal tract and respiratory organs, the spinal cord, the skeletal system, and the hematopoietic system develop simultaneously. system. Not surprisingly, this important developmental period in mice has been intensively studied. Indeed, most of the key regulators of developmental defects can be studied during this time period ^5,6 .

[00315] A common paradigm for the study of organogenesis in mice involves focusing on a single organ system at a limited developmental stage and combining gene knockout studies with anatomical morphology phenotyping, in situ hybridization, immunohistochemistry ^7,8 or, more recently, transcriptome or epigenome profiling ⁹ . Although such targeted studies have provided fundamental insights into mammalian development, however, current technologies do not have sufficient bandwidth and resolution to provide general information about the dynamic molecular processes that occur in diverse and rapidly growing populations and subpopulations of cells during organogenesis.

[00316] "Shotgun profiling", referring to the molecular content of individual cells, is a promising avenue to address these issues and further advance understanding of the mammalian developmental pathway. For example, single cell RNA-seq sequencing techniques have recently revealed high heterogeneity in neurons and myocardiocytes during development in mice ^10,11 . Although two transcription atlases of individual mouse cells have recently been published and provided important resources for the field, ^12,13 they are mostly limited to adult organs and no attempt has yet been made to characterize the emergence and temporal dynamics of mammalian cell types during their development.

[00317] Single cell combinatorial indexing ("sci-") is a methodological framework that uses split-merge barcoding to uniquely label the composition of nucleic acids in a large number of single cells or nuclei ^14,21 . Recently, a "sci-protocol" for transcriptomes ("sci-PHK-seq") has been developed by the authors, and this protocol has been applied to create a 50-fold "cell coverage" by the "shotgun" method for the L2 stage nematodes Caenorhabditis elegans ¹⁹ . Although the throughput of sci-'-methods increases exponentially with increasing number of indexing rounds, however, this potential has yet to be fully realized with respect to other factors such as the rate of cell loss and limited reaction efficiency at some steps ^19,21 . To solve this problem, the authors developed and carefully optimized the 3-level sci-PHK-seq (sci-PHK-seq3) method, resulting in a working protocol that allows profiling over one million cells per experiment. As shown previously ¹⁹ , multiple samples (eg, cues, time points, etc.) can be barcoded during the first round of indexing and processed simultaneously.

[00318] In this case, the authors attempted to investigate the transcriptional developmental dynamics of mice during organogenesis in single cell resolution using sci-RNA-seq3. In one experiment, the authors profiled over 2 million individual cells obtained from 61 mouse embryos between E9.5-E13.5 (10-15 repeats at each time point). Based on these data, the authors identified 38 major cell types, as well as more than 600 more granular cell types (referred to here as "subtypes" to distinguish them from the 38 major cell types). In general, the authors have discovered thousands of new candidate genes for certain cell types and subtypes, which were confirmed by the authors as representative examples of their overall level of in situ hybridization. The authors quantified the dynamics of proliferation and gene expression in proliferating and transient cells in the second trimester of pregnancy, including targeted analyzes of only the apical ectodermal ridge, limb mesenchyme, and skeletal muscle. Using the new algorithm, the authors were

the main developmental pathways of individual cells during murine organogenesis have been identified and examples of different pathways to reach the same endpoint, ie, branching and convergence, have been found. These data are freely available and are presented in a way that makes it easier to understand the information constantly commented on by scientists.

[00319] Results

[00320] Profiling 2 million cells from 61 mouse embryos at 5 developmental stages with sci-RNA-seq3

[00321] To increase the throughput of sci-PHK-seq, over 1000 experimental conditions were investigated by the authors. Regarding the authors' original description of this method ¹⁹ , the main improvements introduced with sci-PHK-seq3 (Fig. 4A, Methods) are that: (i) The authors developed a new strategy for isolating and fixing nuclei, where nuclei isolated directly from fresh tissue without any enzymatic treatment. The nuclei, after being isolated, are fixed in 4% paraformaldehyde and these nuclei can be stored in liquid nitrogen until the next treatment (ii). Compared to the previous description of 3-level indexing ¹⁹ , the authors have moved from indexed Tn5 labeling to indexed hairpin ligation. (iii) Several individual reactions, such as the reverse transcription reaction, were further optimized for efficiency. (iv) No FACS sorting step was performed, and sonication and filtration steps were performed to minimize nuclear aggregation. The sci-PHK-seq3 library preparation steps can be completed by one person in one week, and far exceed the "experimental" throughput of alternative sc-RNA-seq protocols (FIG. 4B).

[00322] We harvested C57BL/6 mouse embryos between E9.5-E13.5, and flash frozen in liquid nitrogen, including 10-15 embryos from at least three independent litters per stage. We subsequently isolated nuclei from 61 single whole embryos and performed sci-RNA-seq3 (FIG. 4A). The nuclei obtained from each embryo were placed in different wells during the first round of indexing so that the RNA-seq profiles of the individual nuclei could match the embryos from which they were obtained (FIG. 5A). As an internal control, the authors also added a mixture of HEK293T and NIH/3T3 cells to two wells during the first round of indexing. After completion of the sci-PHK-seq3 protocol, the resulting library was sequenced in one round of NovaSeq to obtain 11 billion reads (Fig. 5B).

[00323] From this single experiment, we generated 2,072,011 single cell transcriptomes (unique molecular identifier number or UMI number ≥200), including 2,058,652 cells from 61 mouse embryos and 13,359 cells derived from HEK293T or NIH/3T3 cells. It can be predicted with certainty that HEK293T and NIH/3T3 cell transcriptomes are overwhelmingly mapped to one species or another species genome with 420 (3%) collisions (Fig. 4C). At a sequencing depth of 23207 reads per cell, a median of 3676 UMI for HEK293T cells and 5163 UMI for NIH/3T3 cells was observed, with 3.9% and 2.9% of reads per cell mapped to mismatched species, respectively (Fig. 5C- D). The authors compared the original sci-PHK-seq protocol with sci-PHK-seq3 by reducing the sample size of the previously acquired dataset ¹⁹ to an equivalent sequencing depth per HEK293T or NIH/3T3 cell. The sci-RNA-seq3 protocol produced a 40-fold increase in throughput and demonstrated comparable efficiency in terms of the number of UMIs detected per cell (FIG. 5E). In addition, the pooled transcriptome profiles of individual HEK293T cells generated according to the sci-PHK-seq3 and sci-PHK-seq protocols were highly correlated (Pearson: 0.98, Fig. 5F).

[00324] The 2,058,652 embryo-derived cells were mapped to 61 individual embryos based on their first round barcoding (median 35,272 cells per embryo; FIG. 4D). The number of cells isolated from each embryo correlated well with the number of wells in the first round corresponding to those cells (Spearman: 0.75, Fig. 6A). At a relatively shallow sequencing depth (~5000 reads per cell), a median of 519 genes (671 UMI) per cell was determined (FIG. 4E). This result was comparable to or superior to other scRNA-seq studies in which various cell types were isolated and annotated ^19,21,22 despite the fact that the number of initial sequencing reads per cell was less than one third (Fig. 6B-D). Embryos at a later stage (E12.5 and E13.5) had slightly lower UMI per cell, indicating a decrease in mRNA content per nucleus during development (Fig. 6E).

[00325] Based on crude estimates of the number of cells per embryo at each time point (Methods) and the summation of all 10-15 replicates at each time point, the authors performed a shotgun cell coverage estimate for mouse embryos, and this estimate was 0.8 x per E9.5 (200K cells per embryo; in this case 152K profiled), 0.3x per 10.5 (1.1M cells; 378K profiled), 0.2x per E11.5 (2M cells; profiled 616K), 0.08 at E12.5 (6M cells; 475K profiled) and 0.03x at E13.5 (13M cells; 437K profiled). Thus, even before the authors reduced the sample size, it was found that the number of cells that were profiled at each stage is equivalent to a significant percentage of the number of cells in individual mouse embryos (3-80%).

[00326] To check the quality of the data, the transcriptomes of individual cells from each individual were pooled, resulting in 61 “pseudo-volume profiles” of mouse embryos. After counting the number of UMIs mapped to the Xist transcript (expressed only in females) or transcripts on the Y chromosome, mouse embryos could be easily divided into male (x=31) and female (n=30) groups (Fig. 4F) with balanced representation in terms of the number of male and female replicates at each stage (Fig. 6F).

[00327] For additional quality control, the authors performed t-stochastic neighboring region (t-SNE) insertion into the "pseudo-bulk" transcriptomes of 61 embryos, resulting in five closely clustered groups that perfectly matched their developmental stages (Fig. 7A). The authors also ordered mouse embryos along a "pseudo-temporal trajectory" using Monocle ²³ based on the top 1000 genes differentially expressed at different time points, and the resulting ordering was also as expected (Fig. 4G). At the same time, two noticeable gaps in the “pseudo-temporal trajectory” at the level of embryos were observed, that is, one gap between E9.5 and E10.5, and the other between E1 1.5 and E12.5, which suggests the presence of abrupt changes in the global transcriptome in these windows. For each embryo, the authors assigned a pseudo-time that potentially reflected a more detailed assessment of developmental stage (FIG. 7B). Thus, for example, E10.5 embryos at an earlier stage of development were morphologically different from embryos at a later pseudotemporal stage of development (Fig. 7C).

[00328] The authors also examined changes throughout the transcriptome during its development. 12,236 genes were differentially expressed at different developmental stages (data not shown) and the authors plotted the data for some of the most dynamic genes as shown in FIG. 4H. As expected, there was an increase in expression of adult hemoglobin genes such as Hbb-bt and Hbb-bs and a decrease in expression of fetal hemoglobin genes such as Hbb-bhl and Hbb-x. Genes with a known role in neuronal differentiation, including Cntn4 ²⁴ , Neurod2 ²⁵ and Neurod6 ²⁶ , were upregulated at later stages. However, most of the most highly dynamic genes, such as Slc35f4, Prtg and Trim30a, have not been previously characterized. Irrespective of this, and indeed, the authors suggest, the motivation for collecting data on single cells is that the dynamics of gene expression of the "whole embryo" is mainly due to sharp changes in the relative number of cells of various types, but not changes within the cells of any one type.

[00329] Identification and annotation of major cell types and subtypes present during organogenesis in mice

[00330] To identify major cell types, we performed Leuven clustering on 2,058,652 individual cell transcriptomes (i.e., all embryos from all time points), resulting in the identification of 40 different groups followed by t-SNE visualization (FIG. 8A). However, it is reassuring that although there were clear differences between cells obtained at different time points (Fig. 9A), however, cells obtained from replicated embryos at the same time point were distributed in the same way (Fig. 10 ). Based on sets of genes specific to each of these 40 clusters, the authors manually indexed cells by their types by comparison with published marker genes (data not shown). For 37 clusters, the authors could reliably and accurately assign one letter to each cell, while two clusters corresponded to a specific erythroid lineage. One cluster had abnormally high levels of UMI but a low number of cluster-specific genes, suggesting that this may be a technical artifact of cellular doublets. Pooling certain clusters of the erythroid lineage and discarding this putative doublet cluster revealed 38 major cell types (Fig. 8A). For many clusters, highly specific marker genes allow precise differentiation of cells according to their type (Fig. 8B, Fig. 9B-C, data not shown). Thus, for example, cluster 6 (epithelial cells) specifically expressed the well-characterized Epcam and Trp63 marker genes ^21,28 while cluster 29 (hepatocytes) was specifically marked for Afp and Alb ¹² expression. Smaller clusters, including some clusters corresponding to highly specialized cell types, can also be easily annotated. For example, cassette 36 was enriched in transcripts highly expressed during retinal development, such as Tyr and Trpm1, strongly suggesting that these cells are melanocytes ^{29, 30} . Cluster 37 was enriched in transcripts exclusively expressed in the developing lens. For clusters corresponding to embryonic mesenchymal and connective tissue, the identification of specific cell types has proven to be more difficult, mainly because a smaller number of highly specific marker genes are indicated in the current literature.

[00331] Of 26,183 genes, 17,789 genes (68%) were differentially expressed (FDR 5%) in 38 major cell types (FIG. 9B, data not shown). Of these cells, the authors identified 2863 cell-specific markers, the vast majority of which, to the authors' knowledge, were not previously associated with the corresponding cell type (average 75 markers per cluster; Fig. 8B, Fig. 9C). As an example of how these data will be useful for identifying new markers of gene expression specific to developmental stages and cell types, the sonic hedgehog (Shh) gene, which has been shown to play an important role in the development of many organ systems, was considered. including limbs, midbrain structures, thalamus, spinal cord and lungs31. The authors detected the highest expression of Shh in cluster 30 (notochord; data not shown), along with Ntnl, Slit1, and Spon1, all of which are known to be expressed in cells of the notochord and in the ventral plate of the neural tube during development ^32-34 . However, the genes Tox2, Stxbp6, Schip1, Frmd4b, not previously described as notochord markers, were also highly specific to cluster 30.

[00332] As expected, we observed marked changes in cell type ratios during organogenesis. While most of the 38 major cell types proliferated exponentially, some were transient and finally disappeared at E13.5 (Fig. 11A-B). For example, a primitive erythroid lineage derived from the yolk sac represented by cluster 26 was characterized by Hbb-bhl expression, and a definitive erythroid lineage derived from the fetal liver was labeled by Hbb-bs expression in cluster 22 (data not shown). are given). At the E9.5 stage, the authors found mainly cells corresponding to the primitive erythroid lineage (Fig. 8A). Over the next 5 days, the definitive erythroid lineage became the predominant cell type in the fetal circulation and eventually became exclusively the erythrocyte lineage at E13.5 (Fig. 8A). The corresponding gene markers showed similar dynamics (Fig. 11C).

[00333] For the 38 major cell types identified here, the median is 47,073 cells, with the largest cluster containing 144,648 cells (connective tissue precursors; 7.0% of the total data set) and the smallest cluster containing only 1,000 cells (monocytes/granulocytes ; 0.05% of the total data set). Since heterogeneity of certain cell types was quite evident in many of these 38 clusters, the authors adopted an iterative strategy involving Leuven re-clustering for each major cell type in order to identify subclusters (FIGS. 12-13). After removing subclusters dominated by one or two embryos and merging very similar subclusters (Methods), a total of 655 subclusters (referred to here as "subtypes" to distinguish them from the 38 major cell types; Figs. 12-13) were identified. It should be noted that the sensitivity of detection of cells of certain types and subtypes, assessed by the authors in this study, directly depends on the large number of profiled cells. For example, repeating Leuven clustering on 2.5% of the data obtained by the authors (50,000 cells) revealed only a subpopulation of cells of certain types and subtypes (Fig. 14).

[00334] The 655 subtypes consist of a median of 1869 cells and range from 51 cells (notochord cell subtype) to 65894 cells (connective tissue progenitor subtype) (FIG. 15A).

[00335] Nearly all subtypes (99%) are composed of cells from multiple embryos, with no single embryo dominating (Fig. 15B-C). In order to confirm that these subtypes are included in compatible transcriptional programs that differ from related subtypes, a median of 55 specific markers per subtype was determined by the authors (Fig. 15D; it should be noted that subtype-specific markers are defined by specificity within the respective main cell types, not in the general dataset). Thus, for more than 38 major cell types, individual subtypes showed changing dynamics between E9.5 and E13.5. Most of the subtypes (64%) showed an increase in the estimated number of cells, while 12% of the cells had a reduced number, and 24% showed more complex patterns (Fig. 16A-B). Interestingly, the authors could easily separate embryos at different developmental stages based solely on the ratio of cells corresponding to each subtype (Fig. 16C).

[00336] Characterization of gene expression trajectories during limb apical ectodermal ridge (AER) development

[0 0337] As an example of an organ that can be represented with detailed annotations and subtype studies, the authors chose the epithelium (cluster 6), and in particular the apical ectodermal ridge (subcluster 6.25). Based on subtype-specific marker genes, the authors annotated 29 epithelial subtypes (cluster 6; Fig. 17A; Fig. 18A, data not shown). For example, epithelial cells of subtype 6.10 were labeled with Oc90, i.e., a gene expressed exclusively in the otic vesicle epithelium ³⁵ , while epithelial cells of subtype 6.25 showed increased expression of the well-characterized marker genes Fgf8, Msx2, and Rspo2 specific to the apical ectodermal ridge ( AER), that is, a highly specialized epithelium involved in digit development ³⁶ . For all epithelial subtypes, the authors identified genes not previously known as markers. For example, AER also differed in the expression of Fndc3a, Adamts3, Slc16a10, Snap91, and Pou6f2. Common in situ hybridization (WISH) of Fgf8 (a known marker) and Fndc3a (a novel marker) confirmed that both of these genes are expressed at the most distal end of the renal limb nodes representing the AER at E10.5 (Fig. 17B-E).

[00338] Next, the authors investigated the dynamics of cell proliferation and gene expression during the development of AER. The authors identified a total of 1237 AER cells, which was only 0.06% of the total data set obtained by the authors, but they were involved in the development of almost every embryo (45 out of 61 with more than 5 profiled AER cells). Although AER cells were detected at all time points, however, as observed by the authors, they were at their peak in terms of the number of cells per embryo at the E9.5 stage, and then decreased (Fig. 17F), which was consistent with previous reports ³⁷ and conducted by the authors in situ validation studies (FIG. 17C). To characterize the dynamics of AER gene expression during development, we performed a pseudotemporal ordering of AER cells based on the best 500 differentially expressed genes during developmental stages, which allowed us to create a simple trajectory of early or late development (Fig. 17G). 710 protein-coding genes were differentially expressed by developmental pseudotime (FDR 5%) (data not shown). Thus, for example, Fgf9, which is known to have AER-specific expression in the renal nodes of the extremities ³⁸ , showed a slower activation dynamics compared to Fgf8 and Fndc3a (Fig. 17H). Largely activated genes may play an important role in AER cell differentiation. Thus, for example, activated genes include Rspo2, which is known to play an important role in maintaining AER and in maintaining growth and patterning in limb development ³⁹ (Fig. 17H).

[00339] We also identified genes that were significantly downregulated in AER cells between E9.5 and E13.5 (169 genes at 1% FDR; FIG. 19A). These genes are Ki67 (MM67) and insulin-like growth factor 2 (Igf2), both of which play a role in stimulating cell proliferation ^40,41 (Fig. 17H). Indeed, consistent with the cessation of AER proliferation in this developmental time window, analyzes of the level of the pathway to significantly downregulate gene expression revealed timing associated with cell cycle progression and glucose metabolism, as well as transcription factors associated with pluripotency (Is11, Pou5f1, Nanog) (Fig. .19B-C).

[00340] Characterization of trajectories of cell fate change during the development of limb mesenchyme

[00341] The authors then attempted to explore the developmental pathways that cells of different types undergo during this important period of mammalian development, including transitions of cells from one type and subtype to another. Most modern pseudo-time trajectory reconstruction algorithms have two main limitations. First, when creating them, it is assumed that the cells are in one continuous variety, that is, without gaps between subpopulations of cells. However, since the earliest embryos according to the invention are formed at the E9.5 stage, the data set obtained by the authors does not contain cells corresponding to at least some of the hereditary conditions. Second, these algorithms assume that the underlying trajectory is a tree, with branching points corresponding to cell fate decisions. However, some tissues are known to contain transcriptionally indistinguishable cells brought about by transcriptionally distinct lineages, ie convergence of trajectories separated by one or more branching events.

[00342] To overcome these limitations, the authors have developed a new algorithm, included in the Monocle ⁴² package, in order to allow multiple non-intersecting paths, as well as to provide both branching and convergence within the paths. The Monocle 3 algorithm starts by projecting the cells into a small space corresponding to a transcription state using the Uniform Multiple Approximation and Projection (UMAP) program ⁴³ . Then, using Monocle 3, communities of similar cells are detected using Leuven clustering and neighboring communities are combined using statistical analysis included in the Approximate Graph Compression ⁴⁴ (AGA) algorithm. It is important to note that these procedures allow the maintenance of many non-overlapping cell communities. The last stage in Monocle 3 is aimed at resolving the developmental paths of individual cells and identifying the locations of not only branches but converging elements in the set of cells that make up each community, that is, trajectories. Previously, the authors described a procedure called "L1-graph construction" for embedding a "basic graph" in the projection of individual cell PHK-seq profiles, so that each cell is located near a certain point on the graph ⁴⁵ . Although it is possible to study trajectories with closed loops and branches using the L1 graph, however, it could only work on datasets with several hundred cells. In order for the algorithm to process thousands or even millions of cells, two improvements were made by the authors. First, the authors ran this algorithm on several hundred data centroids, not on the cells themselves. Second, they restricted the linear programming procedure of the algorithm to respect the boundaries between non-overlapping trajectories defined by the AGA test.

[00343] We first attempted to apply this novel algorithm to one major cell type, cluster 25, 26559 cells of which were annotated as limb renal ganglion cell mesenchyme for expression of Hoxd13, Fgf10 and Lmx1b (data not shown). Visualization of the cell trajectory of this cluster using Monocle 3 illustrates the dramatic multiplication of mesenchymal limb cells during development, with the main growth observed between stages E10.5 and E12.5 (Fig. 20A). Gene expression can be anything but constant during such breeding, with the levels of 4763 protein-coding genes changing significantly (FDR 1%; data not shown). Early stages of limb mesenchymal development are characterized by some putative genes such as Tbx15 ⁴⁶ and Gpc3 ⁴⁷ and later stages are characterized by Msx1 ⁴⁸ , Epha4 ⁴⁹ and Dach1 ⁵⁰ (Fig. 21A), but the vast majority of dynamically expressed genes are new. Transcription factors significantly upregulated during limb mesenchymal development included factors involved in chondrocyte differentiation (eg, Sox9 ⁵¹ and Yap1 ⁵² ), muscle differentiation (eg, Tead4 ⁵³ ), and in wound healing and limb regeneration (eg, Smarcd1 ⁵⁴ ) ( Fig. 21B).

[00344] Interestingly, forelimb and hindlimb cells were not easily separated by uncontrolled clustering (Fig. 21C) or trajectory analysis (Fig. 22A), but could be distinguished by mutually exclusive Thx5 expression in forelimbs (2085 cells, 7, 9% of all limb mesenchymal cells) and Pitx1 in the hind limbs (1885 cells, 7.1% of all limb mesenchymal cells), with only 22 cells expressing both markers (0.08% of all limb mesenchymal cells compared to ~ 0.6% of expected cells if they were independent Fig. 20B) ⁵⁵ . Thus, 285 genes were differentially expressed in forelimb and hindlimb cells (Fig. 20C, data not shown). Known marker genes such as Tbx4 and Hoxc cluster genes (Hoxc4-10) ⁵⁶ were upregulated in hindlimb cells as expected, but new markers were also identified. For example, the authors found that Epha3 and Hs3st3bl had a 5-fold excess in the forelimbs, while Pcdh17 and Igf1 had a 3-fold excess in the hind limbs.

[00345] Although developmental time is the main axis of limb mesenchymal trajectory variation in the Monocle 3 algorithm (FIG. 20A), there is obviously additional structure. At least some of them seem to correspond to the two main spatial axes of limb development: the proximal-distal axis (the main direction of growth) and the anterior-posterior axis (corresponding to five fingers) ⁵⁵ . Thus, for example, Sox6 and Sox9 (proximal) 57, 58, Hoxd13 and Tfap2b (distal) 36, Pax9 and Alx4 (anterior) and Shh and Hand2 (posterior) were differentially distributed along the trajectory according to Monocle 3 (Fig. 20D; Fig. 22V). In situ hybridization of the entire line Hoxd13 (a known distal marker) and Cpa2 (a novel marker whose distribution along the trajectory according to Monocle 3 was similar to the distribution of known distal markers) confirmed that both genes are expressed in the mesenchyme of distal limbs at stages between E10.5 and E13 .5 (FIGS. 20F-H). Application of the Moran's test to determine the spatial autocorrelation to the limb mesenchymal trajectory revealed 1191 significantly different genes (FDR 1%; Moran's I>10). These genes were grouped into eight expression patterns, some of which coincide with the distribution of markers on the proximal-distal and anterior-posterior axes (Fig. 23, data not shown).

[00346] The combined summary results obtained by the authors for AER trajectories and limb mesenchyme are shown in FIG. 20I. Although limb development follows relatively simple trajectories, the authors' analyzes show how this atlas of individual cells involved in mouse organogenesis can be used to characterize the spatiotemporal dynamics of gene expression in specific systems.

[00347] Definition and differentiation characterization of major cell lines of mouse organogenesis

[00348] Next, the authors attempted to identify major developmental lineages and cellular trajectories for the entire data set. Using the Monocle 3 algorithm, 100,000 high quality cells (UMI>400) were sampled across eight well-defined lineages (Fig. 24A, Fig. 25A). Nearly all of the 38 major cell types almost exclusively fall into one of these eight groups (FIG. 24B). The exceptions are three of the four smallest clusters: monocytes/granulocytes (36 cells), lens (125 cells) and megakaryocytes (287 cells), probably due to their small number. The two most complex structures appear to be the mesenchymal trajectory, which includes all types of mesenchymal and muscle cells (left, Fig. 24A and Fig. 25A), and the neural tube/notochord trajectory, which includes the notochord, neural tube, progenitor cells, and developing neurons and glial cells (right, Fig. 24A and Fig. 25A). The first neural crest trajectory ("neural crest 1") includes melanocytes and Schwann cell precursors, and the second neural crest trajectory ("neural crest 2") consists of sensory neurons. The hematopoietic trajectory includes megakaryocytes, erythrocytes, and lymphocytes, while each of the other three trajectories (hepatic, endothelial, epithelial) correspond to one major cell type. Although the estimated number of cells per embryo in each of these lineages increases exponentially from stage E9.5 to E13.5, however, their ratios remain relatively stable, with the exception of hepatocytes, which increase almost tenfold in this developmental time window ( 0.3% in step E9.5→2.8% in step E13.5) (Fig. 25B-C).

[00349] UMAP allows cells of the same type to be projected to specific areas, but unlike t-SNE, it also allows cells of related types to be placed next to each other. Thus, for example, early mesenchymal cells appear to diverge from a defined area into myocytes, limb mesenchyme, chondrocytes/osteoblasts, and connective tissues (FIG. 24A, left). Similarly, cell types found later in development, such as glutamatergic neurons, are separated from early CNS progenitors (eg, radial glial cells) by a 'bridge' of neural progenitor cells (Fig. 24A, right). On the other hand, breaks (eg, between eight major lineages) probably indicate the absence of intermediate or hereditary conditions between these groups due to limitations in the authors' studies at stages E9.5-E13.5.

[00350] After the authors separately analyzed each of the eight major lineages using the trajectory evaluation analysis as described above, i.e., by analogy with iterative subclustering, the mesenchymal and neural tube/notochord trajectories were plotted again as described above (FIGS. 26-27, top row), and other major lineages (epithelial, endothelial, etc.) showed many discontinuous sublines, allowing detailed subtype trajectories to be revealed (FIGS. 26-27, remaining rows). Thus, for example, if the authors annotated an epithelial trajectory with 29 subtypes (Fig. 17A), they observed several distinct subtrajectories, each derived from a core concentration of E9.5-derived cells underlying epithelial cell subpopulations of certain subtypes. (Fig. 24C, Fig. 28). Thus, for example, epithelial cells of the apical ectodermal ridge (FIG. 17G) form a linear subtrajectory from cells at stages E9.5-E13.5 that are clearly separated from other epithelial subtrajectories (FIG. 24C, lower center).

[00351] Reconstruction of cell trajectories during skeletal myogenesis

[00352] More research is needed to fully elucidate the relationship between the cell types and subtypes that make up the trajectories shown in FIG. 24 in particular are more complex. As a representative example, the authors performed a more detailed assessment of the development of muscle tissue, which consists of various mesodermal lineages that form before the onset of organogenesis. Thus, for example, the extraocular muscles are derived from the prechordial mesoderm, while the other muscles of the face and jaw are derived from the nasopharyngeal mesoderm. Skeletal myogenesis is shaped by a core set of myogenic regulatory factors (MRFs) that are activated by various sets of upstream genes ⁵⁹ . Thus, for example, Pax3 activates Myod1 in the muscles of the trunk, while in the head region, Pax3 is optional and MRFs activate Pitx2 and Tbx1 ^60-62 . Myod1 or Myf5, in turn, activate myogenin, which triggers the expression of many genes necessary for skeletal muscle contractility. The authors hypothesized that the myogenic trajectory, when viewed on the scale of the entire embryo, would have multiple entry points that feed cells into a common path corresponding to the activation of the nuclear gene expression program common to myotubes.

[00353] To test this hypothesis, the authors isolated myocytes and their putative "ancestor" cells from the mesenchymal pathway, first by quantifying the fraction of cells at each node of the main graph, which were classified as myocytes (cluster 13). Then, all the nodes of the "most myocytes" were collected by the authors, and the edges of the main graph were used to expand this set of nodes into a wider "neighborhood" of cells (Fig. 29A). Then, the authors re-ran Monocle 3 on this cell subset to build a myogenesis-specific trajectory. This trajectory was characterized by multiple focal concentrations of E9.5 cells, with cells in later stages being distributed along several outward pathways (FIG. 29B). Pax3 and Pax7, which are characteristic of skeletal muscle progenitors, were expressed in cells distributed over a wide band of the main graph (Fig. 29C). Two parallel linear segments emanate from this region of the graph, on which cells express either Myf5 or Myod. Both pathways enter the common area occupied by cells expressing Myog or Myh3, myocyte and myotube markers, respectively. An additional pathway taken by cells from the E9.5 stage, which express Lhx2, Tbx1 and Pitx2 but very low levels of Pax3, enters the trajectory just before the Myf5 and Myod1 segments, possibly corresponding to the pharyngeal mesoderm. Thus, the dynamics of MRFs and their upstream activators along the entire trajectory is consistent with the notion that different mesodermal lineages use different factors to align with the underlying muscle gene program (Fig. 29D).

[00354] Discussion

[00355] In this study, the authors attempted to characterize mammalian development by profiling individual cell transcriptomes at the scale of the entire mouse embryo, focusing on a window corresponding to classical organogenesis. By profiling over 2,000,000 cells from 61 individual embryos in a single sci-RNA-seq3 experiment, the authors also developed a technical framework for small laboratories to generate single cell RNA-seq datasets with unprecedented throughput. To resolve branching, convergence, and discontinuities in developmental trajectories, the authors introduced Monocle 3, a new algorithm for trajectory determination that will scale to millions of cells.

[00356] For mouse embryos with an average gestational age, the authors identified 38 major cell types and more than 600 subtypes. Each of these types and subtypes is characterized by the expression of a number of marker genes, the vast majority of which are new, and representative examples of which the authors confirmed by common in situ hybridization. To illustrate the effectiveness of deep cell trapping with the shotgun method for characterizing rare cell types, the authors isolated markers and dynamically expressed genes in the apical ectodermal ridge (AER), a specialized epithelium important in digit development, but only 0.06 are represented here. % profiled cells. The 38 major cell types are broadly classified into 8 trajectories, including the mesenchymal trajectory, neural tube/notochord, hematopoietic, hepatic, endothelial, epithelial trajectory, and two neural crest trajectories. The gap between these eight trajectories is likely due to the lack of representation of ancestral or intermediate states in the authors' data set since E9.5. Analysis of the trajectory of the limb mesenchyme revealed correlates of developmental heterogeneity corresponding to both temporal and multiple spatial axes. Focusing on the trajectory of the subpopulation of mesenchymal cells corresponding to myocytes and their progenitors, the authors identified many subtrajectories that converge into one common endpoint corresponding to myotubes. This example of "convergence" of expression programs differs from the branching structure assigned by most algorithms to determine the developmental trajectory.

[00357] The authors' study has several limitations that must be considered. First, as with other single cell atlases, single cell transcriptome data are scarce. However, previous studies have shown that transcription programs can easily differ in individual cell transcriptome datasets at surprisingly shallow sequencing depths ⁶³ . The ability to identify 655 transcriptionally distinct subtypes, with a median of 671 UMI per cell, is consistent with this notion, and pooling the transcriptomes of each cell type or subtype allows the authors to generate representative expression profiles. Secondly, although the authors are quite confident in most of the assignments to certain types of cells made here, they should nevertheless be considered as preliminary. The key problem is that the development of mice at mid-gestation (E9.5-E13.5) has not previously been studied either in terms of single cell resolution or on an organism scale. Existing transcription atlases for single cells include representative single organs of adult or late embryonic mice ^12,13 . Although the authors have made significant progress to date, however, a complete annotation of these 655 cell subtypes is an ongoing project and the authors expect to receive support from researchers and experts in the field to reach a stable consensus. To this end, the authors have created a public database to facilitate annotation for authors and researchers (available online at atlas.gs.washington.edu/mouse-rna/). The unique page for each subtype includes a loading matrix of the cells that make it up, a list of marker genes specific to that subtype, and a description of the cell dynamics of that subtype in the developmental window studied here.

[00358] It is possible that, ultimately, the distant goal of scientists in this field from a technical perspective is to create a comprehensive, spatiotemporal molecular atlas of mammalian development in single cell resolution. To achieve this goal, the choice was made in mice, which has several advantages, including the small size of the mouse, the availability of early time points of its development, the innate genetic background, and genetic manipulation. These data, representing a cell count corresponding to a significant percentage of the cellular content of an individual mouse embryo (from 3 to 80% "shotgun cell coverage" per stage), are a powerful resource in the field of developmental biology, and may also further advance the development of computer-assisted methods for resolution and interpretation of cell types or developmental trajectories. Looking to the future, the authors expect that a summary assessment of the transcriptome, additional molecular phenotypes ⁶⁴ , lineage histories ⁶⁵ , and a wealth of information will help form a general understanding of mammalian development.

[00359] In conclusion, wild-type mouse single cell developmental atlases are also a first step towards organism-wide understanding of pleiotropic developmental disorders, as well as a detailed study of the subtle role of genes and regulatory sequences in development. For example, although ~35% of gene knockouts in mice are lethal ⁵ , however, many knockouts, and in particular knockouts of conserved regulatory sequences, do not show any abnormalities in conventional phenotyping66. The authors suggest that organism-scale sc-RNA-seq allows for reverse genetics techniques, for example, potentially detecting previously missed phenotypes with subtle defects in molecular programs or the relative ratios of certain cell types ⁶⁷ .

[00360] Methods

[00361] Isolation of embryos

[00362] C57BL/6 mice were obtained from Jackson Laboratories (Bar-Harbor, ME) and paired for mating. The day of mating was considered embryonic day (E) 0.5. Excisions were performed as previously described ⁶⁹ and all embryos were immediately frozen in liquid nitrogen. All animal procedures were performed with the permission of Institute officials, State and Government Regulatory Authorities (in accordance with IACUC protocol 4378-01).

[00363] General in situ hybridization

[00364] mRNA expression in E9.5-E11.5 mouse embryos was assessed by total in situ hybridization (WISH) using a digoxigenin labeled antisense riboprobe transcribed from cloned gene-specific probes (DIG PCR Probe Synthesis Kit, Roche). Whole embryos were fixed overnight in 4% PFA/PBS. Embryos were washed in PBST (0.1% Tween) and dehydrated stepwise in 25%, 50% and 75% methanol/PBST and finally stored at -20° C. in 100% methanol. The WISH protocol was performed as follows: Day 1) Embryos were rehydrated on ice in the reversible steps of methanol/PBST treatment, washed in PBST, bleached in 6% H202/PBST for 1 hour, and washed in PBST. Embryos were then treated in 10 μg/ml proteinase K/PBST for 3 minutes, incubated in glycine/PBST, washed in PBST, and finally re-fixed for 20 minutes with 4% PFA/PBS, 0.2% glutaraldehyde, and 0 .1% Tween 20. After additional washing steps with PBST, embryos were incubated at 68°C in buffer L1 (50% deionized formamide, 5x SSC, 1% SDS, 0.1% Tween 20 in DEPC; pH 4.5) in within 10 minutes. The embryos were then incubated for 2 hours at 68° C. in hybridization buffer 1 (L1 with 0.1% tRNA and 0.05% heparin). Thereafter, embryos were incubated overnight at 68° C. in hybridization buffer 2 (in hybridization buffer 1 with 0.1% tRNA and 0.05% heparin and a 1:500 DIG probe). Day 2) Removal of unbound probe was carried out with a series of washing steps 3×30 minutes each, at 68°C: L1, L2 (50% deionized formamide, 2×SSC, pH 4.5, 0.1% Tween 20 in DEPC; pH 4.5) and L3 (2x SSC, pH 4.5, 0.1% Tween 20 in DEPC; pH 4.5). Embryos were then treated for 1 hour with RNase solution (0.1 M NaCl, 0.01 M Tris, pH 7.5, 0.2% Tween 20, 100 μg/ml RNase A in H ₂ O) followed by washing in TBST 1 (140 mM NaCl, 2.7 mM KCl, 2 5 mM Tris-HCl, 1% Tween 20; pH 7.5). Embryos were then blocked for 2 hours at room temperature in blocking solution (TBST 1 with 2% calf serum and 0.2% BSA) followed by overnight incubation at 4°C in blocking solution containing 1:5000 anti-digoxigenin- AR, 1:5000. Day 3) Removal of unbound antibody was performed with a series of TBST 2 washes (TBST with 0.1% Tween 20 and 0.05% Levamisole/Tetramisole) 8x30 minutes at room temperature and left overnight at 4°C. Day 4) Embryo staining was started by washing at room temperature with alkaline phosphatase buffer (0.02 M NaCl, 0.05 M MgCl ₂ , 0.1% Tween 20, 0.1 M Tris-HCl and 0.05% levamisole/ tetramisole in H ₂ O) 3×20 minutes followed by staining with magenta substrate for AP BM (Roche). Stained embryos were visualized under a Zeiss Discovery V.12 microscope and using a Leica DFC420 digital camera.

[00365] Mammalian cell culture

[00366] All mammalian cells were cultured at 37°C with 5% CO ₂ and stored in DMEM with high glucose content (Gibco, cat. No. 11965) for HEK293T and NIH/3T3 cells, and then added 10% FBS and IX Pen/ Strep (Gibco cat. no. 15140122; 100 U/ml penicillin, 100 µg/ml streptomycin). Cells were trypsinized with 0.25% trypsin-EDTA (Gibco, cat. no. 25200-056) and distributed at a ratio of 1:10 three times a week.

[00367] Isolation and fixation of mouse embryonic nuclei

[00368] Mouse embryos at various stages of development were treated together to reduce the cluster effect. Each mouse embryo was minced into small pieces with a razor blade in 1 ml of ice-cold cell lysis buffer (10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl ₂ and 0.1% IGEPAL CA-630 ⁷⁰ modified so that it includes 1% Super-RNase inhibitor and 1% BSA) and transferred to the top of a 40 µm cell filter (Falcon). Tissues were homogenized with a rubber tip syringe (5 ml, BD) in 4 ml cell lysis buffer. Then, the filtered nuclei were transferred to a new 15 ml tube (Falcon) and pelleted by centrifugation at 500×g for 5 minutes and washed once with 1 ml of cell lysis buffer. The nuclei were fixed in 4 ml ice-cold 4% paraformaldehyde (EMS) for 15 minutes on ice. After fixation, nuclei were washed twice in 1 ml of nuclear wash buffer (cell lysis buffer without IGEPAL) and resuspended in 500 µl of nuclear wash buffer. Samples were dispensed into two tubes, 2 x 50 µl in each tube, and flash frozen in liquid nitrogen.

[00369] For quality control evaluation, HEK293T and NIH/3T3 cells were trypsinized, centrifuged at 300×g for 5 minutes (4°C) and washed once in 1×PBS. An equal number of cells and NIH/3T3 were pooled and lysed using 1 ml of ice-cold cell lysis buffer, followed by fixation and storage under the same conditions as for mouse embryos.

[00370] Obtaining and sequencing the sci-RNA-seq3 library

[00371] Thawed kernels were permeated using 0.2% Triton X-100 (in Kernel Wash Buffer) for 3 minutes on ice and then rapidly sonicated (Diagenode, 12 seconds on low power) to reduce kernel agglomeration . Then, the nuclei were washed once with nuclear washing buffer and filtered through a 1 ml Flowmi cell filter (Flowmi). The filtered nuclei were centrifuged at 500×g for 5 minutes and resuspended in nuclear washing buffer.

[00372] The nuclei from each mouse embryo were then distributed to several individual wells in four 96-well plates. The relationship between the well identifier and the mouse embryo was then recorded for further data processing. For each well, 80,000 nuclei (16 μl) were mixed with 8 ml of 25 μM oligo-dT anchored primer (5'-/5Phos/CAGAGCNNNNNNNNN [10 bp barcode] TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTT-3', (SEQ ID NO: 1 ), where "N" is any base; IDT) and 2 μl of 10 mM dNTP mixture (Thermo) were denatured at 55° C. for 5 minutes and immediately placed on ice. Thereafter, 14 µl of a first strand reaction mixture containing 8 µl of 5× Superscript IV first strand buffer (Invitrogen), 2 µl of 100 mM DTT (Invitrogen), 2 µl of Superscript IV reverse transcriptase (200 U) was added to each well. /µl, Invitrogen), 2 µl of recombinant RNaseOUT ribonuclease inhibitor (Invitrogen). Reverse transcription was performed by incubating the plates in a temperature gradient (4°C, 2 minutes, 10°C, 2 minutes, 20°C, 2 minutes, 30°C, 2 minutes, 40°C, 2 minutes, 50°C, 2 minutes and 55°C, 10 minutes).

[00373] After the RT reaction, 60 μl of nuclear dilution buffer (10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl ₂ and 1% BSA) was added to each well. The nuclei from all wells were pooled and centrifuged at 500×g for 10 minutes. The nuclei were then resuspended in nuclear wash buffer and redistributed into four other 96-well plates, where each well contained 4 µl T4 ligation buffer (NEB), 2 µl T4 DNA ligase (NEB), 4 µl betaine solution (5M, Sigma -Aldrich), 6 µl nuclear wash buffer, 8 µl barcoded ligation adapter (100 µM, 5'-GCTCTG [barcode 9bp or 10bp A]/i-deoxy- U/ACGACGCTCTTCCGATCT [reverse complement of barcode A]-3') (SEQ ID NO: 2) and 16 μl of 40% PEG 8000 (Sigma-Aldrich). The ligation reaction was carried out at 16°C for 3 hours.

[00374] After the RT reaction, 60 μl of nuclear dilution buffer (10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM and 1% BSA) was added to each well. The nuclei from all wells were pooled and centrifuged at 600×g for 10 minutes. Nuclei were washed once with nuclear wash buffer and filtered twice through a 1 ml Flowmi cell filter (Flowmi), counted and redistributed into eight 96-well plates, each well containing 2500 nuclei in 5 μl of nuclear wash buffer and 5 µl of elution buffer (Qiagen). Then, 1.33 µl of second strand mRNA synthesis buffer (NEB) and 0.66 µl of second strand mRNA synthesis enzyme (NEB) were added to each well, and second strand synthesis was performed at 16°C for 180 min.

[00375] For labeling, the contents of each well were mixed with 11 µl of Nextera TD buffer (Illumina) and 1 µl of [i7] TDE1 enzyme only (62.5 nM, Illumina) and then incubated at 55°C for 5 minutes to conduct tagging. The reaction was then terminated by adding 24 μl of DNA binding buffer (Zymo) per well and incubating at room temperature for 5 minutes. Each well was then purified using 1.5x AMPure XP spheres (Beckman Coulter). At the elution step, 8 µl of nuclease-free water, 1 µl of 10x USER buffer (NEB), 1 µl of USER enzyme (NEB) were added to each well and incubated at 37° C. for 15 minutes. An additional 6.5 μl of elution buffer was added to each well. The AMPure XP spheres were removed with a magnetic rod and the eluent was transferred to a new 96-well plate.

[00376] For PCR amplification, the contents of each well (16 µl of product) were mixed with 2 µl of 10 µM indexed primer P5 (5'-AATGATACGGCGACCACCGAGATCTACAC[i5]ACACTCTTTCCCTACACGACGCTCTTCCGATCT-3'; IDT) (SEQ ID NO: 3), 2 µl 10 μM primer P7 (5'-CAAGCAGAAGACGGCATACGAGAT[i7]GTCTCGTGGGCTCGG-3', IDT) (SEQ ID NO: 4) and 20 μl 2x NEBNext High-Fidelity Master Mix (NEB) PCR mix. Amplification was carried out according to the following program: 72°C for 5 minutes, 98°C for 30 seconds, 12-14 cycles (98°C for 10 seconds, 66°C for 30 seconds, 72°C for 1 minute ) and finally at 72°C for 5 minutes.

[00377] After PCR, samples were pooled and purified using 0.8 volume of AMPure XP spheres. Library concentrations were determined using Qubit (Invitrogen) and libraries were visualized by electrophoresis on a 6% TBE-PAGE gel. All libraries were sequenced on the same NovaSeq platform (Illumina) (read 1:34 cycles, read 2:52 cycles, index 1:10 cycles, index 2:10 cycles).

[00378] Processing sequencing reads

[00379] The base query sequences were converted to fastq format using bcl2fastq Illumina and demultiplexed based on the i5 and i7 PCR barcodes using the deML ⁷¹ maximum likelihood demuxing software package with default parameters. Further sequence processing and matrix generation for single cell digitization was similar to sci-RNA-seq ¹⁹ , except that the RT index was combined with the hairpin adapter index, and thus the mapped reads were subdivided into composite cell indices by read demultiplexing using OT-index and ligation index (ED<2, including insertions and deletions). Briefly, the demultiplexed reads were filtered by OT-index and ligation-index (ED<2, including insertions and deletions), and the adapter was removed using the trim_galore/0.4.1 program with default parameters. The truncated reads were mapped to the reference mouse genome (mm10) for mouse embryonic nuclei or the chimeric reference genome of human hgl9 and mouse mm10 for mixed HEK293T and NIH/3T3 nuclei using the STAR/v 2.5.2b ⁷² program with default parameters and gene annotations ( GENCODE V19 for humans; GENCODE VM11 for mice). Uniquely mapped reads were isolated and duplicates were removed using the unique molecular identifier (UMI) sequence, reverse transcription index (RT), hairpin ligation adapter index, and read 2 end coordinates (i.e., reads with identical UMI, RT index, ligation adapter index). and tagging site were treated as duplicates). Finally, the mapped reads were assigned to individual cell indices by subsequent read demultiplexing using the OT index and the hairpin ligation index (ED<2, including insertions and deletions). For the mixed species experiment, the percentage of uniquely mapped reads for the genomes of each species was calculated. Cells with more than 85% UMI assigned to the same species were considered as species-specific cells, and the remaining cells were classified as mixed cells or "collisions". To create digital matrices, the authors calculated the number of chain-specific UMIs for each cell, mapped to the exon and intron regions of each gene using the Python HTseq ⁷³ software package. For multimapped reads, these reads were assigned to the closest genes, unless another interspersed gene fell within 100 bp. to the end of the nearest gene, in which case, the read was discarded. For most analyses, the authors included expected UMIs for intron and exon strands on single cell gene expression templates.

[00380] Whole mouse embryo analysis

[00381] After creating a matrix for counting individual cell genes, each cell was assigned to its original mouse embryo based on the OT barcode. Read maps for each embryo were combined to create a "total RNA-seq" for each embryo. To separate embryos by sex, the authors counted reads mapped to female-specific non-coding RNA genes (Xist) or to chr Y genes (with the exception of the Erdr1 gene, which is present in chr X and chr Y). The embryos were easily separated into a female group (with more reads mapped for Xist genes than for chr Y genes) and a male group (with more reads mapped for chr Y genes than for Xist).

[00382] Pseudo-temporal sequencing of whole mouse embryos was performed using Monocle 2 ⁷⁴ . Briefly, a pooled gene expression template was constructed as described above. Differentially expressed genes under different developmental conditions were identified using the differential GeneTest function of Monocle 2 ⁷⁴ . The top 2000 genes with the lowest q value were used to construct a pseudo-temporal trajectory using Monocle 2 ⁷⁴ . Each embryo was assigned a pseudo-temporal value based on its position on the trajectory tree.

[00383] Cell Clustering, t-SNE Imaging, and Identification of Marker Genes

[00384] A digital template for gene expression was created based on the original sequencing data as described above. Cells with less than 200 UMI were discarded. Subsequent analysis was performed using Monocle2 ⁷⁴ and the python scanpy ⁷⁵ software package. Briefly, a certain number of sex-mapped genes were removed prior to clustering and downsizing. The pretreatment step was carried out in a manner similar to the approach used by Zheng et al. ²² , according to the function "zheng17 recipe" (n_top-genes=2000) in scanpy ⁷⁵ . Dimensional data were first downsampled with PCA (30 components) and then with t-SNE, followed by Leuven clustering on 30 principal components (resolution=1.5). 40 clusters were identified. Then, the authors selected 1000 cells from each cluster, and differentially expressed genes in different clusters were identified using the differential GeneTest function in Monocle 2 ⁷⁴ . The genes specific for each cluster were identified in the same way as described in the literature ⁷⁶ . Clusters were assigned to known cell types based on cluster-specific markers (Table 1). One cluster had an abnormally high number of UMIs, but lacked enough cluster-specific genes to suggest that it may be a technical artifact of cellular doublets and therefore could be deleted. The other two clusters, as it turned out, corresponded to the final lineage of erythroid differentiation and were combined. Consensus expression profiles for each cell type were constructed as described in the literature ⁷⁶ . To identify a gene marker specific for a particular cell type, the authors selected a gene that was differentially expressed in different cell types (FDR 5%, odds ratio analysis) and also maximally expressed in each cell type by at least 2-fold an increase compared to other cell types, which rank second in terms of maximum expression.

[00385] To identify a subcluster, we selected high quality cells (UMI>400) for each major type and performed PCA, t-SNE, Leuven clustering similar to the overall cluster analysis. Highly biased subclusters were filtered out if the majority of cells (>50%) of the cluster originated from a single embryo. Very similar subclusters were pooled if their pooled transcriptomes were highly correlated (Pearson's correlation coefficient >0.95) and if the two clusters were close to each other in t-SNE space. Differentially expressed genes in all subclusters were identified for each major cell type as described above.

[00386] To estimate the number of cells of each type (or subtype), the authors first calculated the proportion of cells of each type in an individual embryo, and then the result was multiplied by the total number of cells calculated for each embryo (E9.5: 200000: E10.5: 1100000 ; E11.5: 2600000; E12.5: 6100000; E13.5: 13000000).

[00387] To identify sex-specific cell types (or subtypes), the authors first calculated the number of cells of each type (or subtype) for males and females at five developmental stages. The ratio specific for cells of each type in males and females was compared with the total ratio of the number of cells in males and females at each stage of development. Next, the authors applied a binomial test in R to identify cell types or subtypes with significant differences between males and females for each cell type (x and n are the number of female cells and the total number of cells of each type at each developmental stage, p is the number of female cells at each stage of development). The p value is converted to a corrected q value by the method of Benjamini and Hochberg using the p correction function in R.

[00388]

[00389] Pseudo-temporal analysis of AER and limb mesenchyme

[00390] Pseudo-temporal sequencing of AER cells, forelimbs or hind limbs was performed using Monocle 2 ⁷⁴ . Briefly, differentially expressed genes at five developmental stages were identified using the differential GeneTest function in Monocle 2 ⁷⁴ . The top 500 genes with the lowest q value were used to construct a pseudo-temporal trajectory using Monocle 2 ⁷⁴ , where the number of UMI per cell was used as a covariate in the construction of the tree. Each cell was assigned a pseudo-temporal value based on its position along the trajectory tree. The pseudo-time smooth change in marker gene expression was determined by introducing the plot_genes_in_pseudotim function in Monocle 2 ⁷⁴ . The cells on the trajectory were grouped according to the method described in the literature ⁷⁷ . Briefly, cells were first clustered in similar positions in pseudo-time using k-means clustering along the pseudo-time axis (k=10). These clusters were subdivided into groups containing at least 50 and no more than 100 cells. Next, the authors pooled the transcriptome profiles of the cells in each group. Pseudo-time gene expression was calculated by the method described in the literature ⁷⁷ . Briefly, genes meeting the significance test (FDR 5%) under various treatment conditions were selected and a natural spline was used to plot genes versus pseudotime, where the average number of genes was included as a covariate. The expression of each gene was subtracted from the lowest expression and then divided by the highest expression. Genes with maximum expression during the first 20% of the pseudo-time were labeled as activated genes. Genes with maximum expression in the last 20% of pseudotime were labeled as repressed genes. Other genes have been labeled as transient genes. Enriched reactome elements (Reactome_2016) and transcription factors (SpEA_2016) were identified using the EnrichR ⁷⁸ software package.

[00391] Conclusion about trajectories defined using Monocle 3

[00392] The Monocle 3 workflow consists of 3 major steps in organizing cells along potentially discontinuous trajectories, followed by optional statistical analyzes to look for genes that differ in expression across all of these trajectories. Monocle 3 also includes visualization packages to help you explore 3D trajectories.

[00393] Uniform Multiple Approximation and Projection (UMAP) Dimension Reduction

[00394] Initially, the Monocle 3 program was used to project data into a low-dimensional space, which facilitates the study of the main graph that describes cellular transitions between transcriptomic states. Monocle 3 works in conjunction with UMAP, a recently proposed algorithm based on Riemannian geometry and algebraic topology for dimensionality reduction and data visualization ⁷⁹ . The imaging quality achieved with this algorithm can compete with the quality achieved with the popular t-SNE (t-stochastic neighbor region embedding) method widely used in single cell transcriptomics. However, if t-SNE is primarily aimed at placing highly similar cells in the same regions of low-dimensional space, then UMAP also preserves distance relationships with a wider range. The UMAP algorithm itself is also more efficient (the complexity of the UMAP algorithm is estimated at O(N) compared to O(Nlog(N)) for t-SNE). Briefly, UMAP first gives a topological view of high-dimensional data with local multiple approximations and combines their local fuzzy simplified set representations. UMAP then optimizes the introduction of the lower dimension by minimizing the cross entropy between the low dimensional representation and the high dimensional representation.

[00395] The computational efficiency of UMAP dramatically speeds up the analysis of mouse embryo data. The authors found that UMAP completes the analysis of a data series for two million cells in 3 hours, while the analysis of t-SNE takes more than 10 hours with 10 centers (in this case, a multicenter bh-t-SNE is used). A few implementation details of this method make UMAP more efficient. The UMAP and t-SNE algorithms have two main stages: first, an intermediate structure is created from a high-dimensional space (usually a space with reduced upper PCA), and then low-dimensional inclusions are searched to represent the intermediate structure. In the second stage, both methods use a stochastic lattice reduction approach with different loss functions to include data in a low-dimensional space. t-SNE requires a loss function for global normalization, while UMAP does not require a loss function but uses a different objective function. This stage allows UMAP to be substantially linearly scaled over a set of data samples. In Monocle 3, the authors work with the Python UMAP implementation program (available online at www.atgithub.com/lmcinnes/umap) from Leland McInnes and John Healy using the reticulate software package (available online at atcran.r-project. org/web/packages/reticulate/index.html).

[00396] Distribution of cells along interrupted trajectories

[00397] Recently, Wolf and colleagues proposed the idea of organizing the collection of transcriptome data for individual cells into an "Abstract Divided Graph" (AGA) linking clusters of cells that can be related to each other in terms of developmental time. Briefly, their algorithm builds a k-nearest neighbor graph on cells and then identifies "communities" of cells in a Leuven method, similar to previous methods for analyzing CyTOF data or collecting RNA-seq data for individual cells ⁸⁰ . Then, using AGA, one can construct a graph in which the vertices are the Leuven communities. Two vertices are connected to an edge in the AGA graph if cells in the respective communities are neighbors in the kNN graph more often than would be expected in a simple binomial model ⁸¹ . Similar methods have also recently been developed and applied to the analysis of cell atlas datasets for zebrafish and frogs ^{82, 83} .

[00398] Monocle 3 relies on the following ideas: first, a kNN graph is built for cells in UMAP space, and then they are grouped into Leuven communities, and each pair of communities is tested for a significant number of connections between their respective cells. Communities that have more connections than expected under the null hypothesis of false association (FDR<10%) remain connected on the AGA plot, while those that fail this test are broken. The resulting AGA graph will have one or more components, each of which is passed to the next step (L1-graph) as a separate group of cells, which will be organized into a trajectory. The AGA algorithm essentially stops at this point, representing the AGA graph as a kind of large trajectory in each community, reflecting the cells in the various states they may take as they develop. In contrast, as described in the next section, Monocle 3 uses the AGA graph to limit the space of the main graphs that can form the final trajectory. That is, Monocle 3 uses a large AGA graph to learn a small trajectory.

[00399] The Monocle 3 implementation of the above procedures scales to millions of cells. Briefly, it uses the Leuven clustering feature from the igraph software package for community discovery. Then, the basic AGA calculations made by Wolf et al. are processed on the computer using a series of sparse matrix operations. X can be taken as a (sparse) matrix representing a member of the cell community. Each X column represents a Leuven community, and each X row corresponds to a specific cell. X _ij =1 if the cells belong to the Leuven community j, otherwise X _ij =0. Next, one can obtain the adjacency matrix A of the kNN graph used to implement Leuven clustering, where A _ij =1 if cell i is connected to j on the kNN graph. Then the matrix M of the connection between each cluster is calculated by the formula:

[00400] M=X ^T AX

[00401] After constructing M, authors may follow Supplementary Note 3.1 of Ref. 81 to calculate the significance of the relationship between each Leuven clustering, and consider any clusters with a p-value greater than the default 0.05 as non-discontinuous.

[00402] Learning the main graph

[00403] Monocle 3 allows you to explore the main graph, which is in the same small-dimensional space as the data to represent the possible paths of cell development. Monocle 3 uses an extended implementation of the ⁸⁴ Ll-graph algorithm to learn the underlying graph. Mao et al. described two variants of the Ll-graph algorithm ⁸⁴ . In the first algorithm ("Algorithm 1"), this data was optimized against all individual input data in the database. Previously, the authors have shown that although L1-graph can be applied to single-cell RNA-seq data, however, it allows the study of the most “noisy” graphs that are not resistant to downsampling, and this algorithm does not effectively scale across datasets for beyond a few hundred cells ⁸⁵ . The authors have not explored "Algorithm 2" described by Qiu et al., where "indicative" input data are first selected using the K-means clustering algorithm. The algorithm is then optimized over this much smaller sample of data. Monocle 3 uses an algorithm that, when applied to cells in the UMAP space, is robust and, with some key changes, can scale to millions of cells.

[00404] The authors' implementation of the L1 graph algorithm has several key features that allow you to analyze large data sets and reliably highlight the main graph. First, the authors studied the L1 graph in UMAP space (by default, in 3-dimensional space). The authors used K-medioids clustering to select "orienting" cells in order to speed up optimization. The number of selected "guide" cells affects the running time of the algorithm and the quality of the solution: too many guides will lead to an unsolvable linear programming problem. Therefore, the authors determined the number of landmarks depending on the data by assigning K to the communities that were found among the cells, and which were three times larger than the Leuven communities, which in practice led to fast and stable solutions.

[00405] The second important optimization of the L1 graph is that the authors impose restrictions on the "feasible" space of all possible graphs W considered in the optimization. Mao et al. considered all possible boundaries between the indicative initial data. However, even with only a thousand "indicative" cells, a linear programming problem can quickly become unfeasible, since the number of variables depends on the number of edges in the graph. In Monocle 3, the authors introduce edges only into the allowed space, which is either in a Minimum Span Tree (MST) built on "guide" points, or a kNN graph (default k=3) built on odd degree vertices in the MST . Finally, the authors excluded edges that would connect cells in different connected components of the AGA graph constructed as described in the previous section.

[00406] Identification of genes whose expression depends on the trajectory

[00407] To identify genes that differ in expression depending on the developmental trajectory, the authors borrowed a statistical criterion commonly used in the analysis of spatial data. Statistical Moran's I-test is a multidirectional and multivariate spatial autocorrelation test. This statistical method allows the determination of spatial relationships between input data through a nearest neighbor graph, making it particularly suitable for analyzing large single-cell RNA-seq datasets.

[004 08] Moran's T-test ⁸⁶ is determined by the formula:

[00409]

представляет собой среднее значение экспрессии генов для ближайших соседей клеток i' (или j'); ω_ij представляет собой весовую матрицу, определенную графом ближайших соседей с нулем по диагонали (то есть, ω_ii=0), и ω_ij=1/k_i, где k_i означает число ближайших соседей, a W равно сумме всех ω_ij.[00410] where N is the number of cells indexed as i and j, and x is the amount of expression of the gene of interest;

represents the average value of gene expression for the nearest neighbors of cells i' (or j'); ω _ij is a weight matrix defined by the nearest neighbor graph with diagonal zero (i.e., ω _ii =0), and ω _ij =1/k _i , where k _i is the number of nearest neighbors and W is the sum of all ω _ij .

[00411] To identify the nearest neighbors used to create the weight matrix W, we first built a nearest neighbor graph k (default 25) (kNN) for all cells in UMAP space. The authors also projected each cell to the nearest node in the main graph. The authors then removed all edges from the kNN graph that connect cells projected onto the nodes of the main graph that are not part of the edge.

[00412] In Monocle 3, the authors implemented the manifoldTest function to identify a set of correlated genes that depend on modified versions of routine programs taken from the spdep software package to implement the Moran's I-test.

[00413] Bibliography

[00414] 1. Fogarty, N. M. E. et al. Genome editing reveals a role for OCT4 in human embryogenesis. Nature 550, 67-73 (2017).

[00415] 2. Kojima, Y., Tam, O. H. & Tam, P. P. L. Timing of developmental events in the early mouse embryo. Semin. celldev. Biol. 34, 65-75 (2014).

[00416] 3. Tam, P. P. L. & Loebel, D. A. F. Gene function in mouse embryogenesis: get set for gastrulation. Nat. Rev. Genet. 8, 368-381 (2007).

[00417] 4. Rivera-Péres J. A. & Hadjantonakis, A.-K. The Dynamics of Morphogenesis in the Early Mouse Embryo. Cold Spring Harb. perspective. Biol. 7, a015867 (2014).

[00418] 5. Dickinson, M. E. et al. High-throughput discovery of novel developmental phenotypes. Nature 537, 508-514 (2016).

[00419] 6. Meehan, T. F. et al. Disease model discovery from 3,328 gene knockouts by The International Mouse Phenotyping Consortium. Nat. Genet. 49, 1231-1238 (2017).

[00420] 7. Shyer, A. E., Huycke, T. R., Lee, C, Mahadevan, L. & Tabin, C. J. Bending gradients: how the intestinal stem cell gets its home. Cell 161, 569-580 (2015).

[00421] 8. Uygur, A. et al. Scaling Pattern to Variations in Size during Development of the Vertebrate Neural Tube. dev. Cell 37, 127-135 (2016).

[00422] 9. Gorkin, D. et al. Systematic mapping of chromatin state landscapes during mouse development. (2017). doi:10.1101/166652

[00423] 10. Mayer, C. et al. Developmental diversification of cortical inhibitory interneurons. Nature 555, 457-462 (2018).

[00424] 11. Lescroart, F. et al. Defining the earliest step of cardiovascular lineage segregation by single-cell RNA-seq. Science (2018). doi:10.1126/science.aao4174

[00425] 12. Han, X. et al. Mapping the Mouse Cell Atlas by Microwell-Seq. Cell 172, 1091-1107.e17 (2018).

[00426] 13. The Tabula Muris Consortium, Quake, S. R., Wyss-Coray, T. & Darmanis, S. Transcriptomic characterization of 20 organs and tissues from mouse at single cell creates resolution a Tabula Muris. (2017). doi:10.1101/237446

[00427] 14. Amini, S. et al. Haplotype-resolved whole-genome sequencing by contiguity-preserving transposition and combinatorial indexing. Nat. Genet. 46, 1343-1349 (2014).

[00428] 15. Adey, A. et al. In vitro, long-range sequence information for de novo genome assembly via transposase contiguity. Genome Res. 24, 2041-2049 (2014).

[00429] 16. Cusanovich, D. A. et al. Multiplex single cell profiling of chromatin accessibility by combinatorial cellular indexing. Science 348, 910-914 (2015).

[00430] 17. Vitak, S. A. et al. Sequencing thousands of single-cell genomes with combinatorial indexing. Nat. Methods 14, 302-308 (2017).

[00431] 18. Ramani, V. et al. Massively multiplex single-cell Hi-C. Nat. Methods 14, 263-266 (2017).

[00432] 19. Cao, J. et al. Comprehensive single-cell transcriptional profiling of a multicellular organism. Science 357, 661-667 (2017).

[00433] 20. Mulqueen, R. M. et al. Scalable and efficient single-cell DNA methylation sequencing by combinatorial indexing. (2017). doi:10.1101/157230

[00434] 21. Rosenberg, A. B. et al. Single-cell profiling of the developing mouse brain and spinal cord with split-pool barcoding. Science (2018). doi:10.112b/science.aam8999

[00435] 22. Zheng, G. X. Y. et al. Massively parallel digital transcriptional profiling of single cells. Nat. commun. 8, 14049 (2017).

[00436] 23. Qiu, X. et al. Reversed graph embedding resolves complex single-cell developmental trajectories. (2017). doi:10.1101/110668

[00437] 24. Fernandez, T. et al. Disruption of contactin 4 (CNTN4) results in developmental delay and other features of 3p deletion syndrome. Am. J. Hum. Genet. 74, 1286-1293 (2004).

[00438] 25. Olson, J. M. et al. NeuroD2 is necessary for development and survival of central nervous system neurons. dev. Biol. 234, 174-187 (2001).

[00439] 26. Uittenbogaard, M., Baxter, K. K. & Chiaramello, A. NeuroD6 Genomic Signature Bridging Neuronal Differentiation to Survival via the Molecular Chaperone Network. J. Neurosci. Res. 88, 33 (2010).

[00440] 27. Yang, A. et al. p63 is essential for regenerative proliferation in limb, craniofacial and epithelial development. Nature 398, 714-718 (1999).

[00441] 28. McQualter, J. L., Yuen, K., Williams, B. & Bertoncello, I. Evidence of an epithelial stem/progenitor cell hierarchy in the adult mouse lung. Proc. Natl. Acad. sci. U. S. A. 107, 1414-1419 (2010).

[00442] 29. Cichorek, M., Wachulska, M., Stasiewicz, A. & Tymińska A. Skin melanocytes: biology and development. Advances in Dermatology and Allergology 1, 30-41 (2013).

[00443] 30. Tomihari, M., Hwang, S.-H., Chung, J.-S., Cruz, P. D., Jr. & Ariizumi, K. Gpnmb is a melanosome-associated glycoprotein that contributes to melanocyte/keratinocyte adhesion in a RGD-dependent fashion. Exp.Dermatol. 18, 586-595 (2009).

[00444] 31. Varjosalo, M. & Taipale, J. Hedgehog: functions and mechanisms. Gene Dev. 22, 2454-2472 (2008).

[00445] 32. Strahle, U., Lam, C. S., Ertzer, R. & Rastegar, S. Vertebrate floor-plate specification: variations on common themes. Trends Genet. 20, 155-162 (2004).

[00446] 33. Holmes, G. P. et al. Distinct but overlapping expression patterns of two vertebrate slit homologs implies functional roles in CNS development and organogenesis. Mech. dev. 79, 57-72 (1998).

[00447] 34. Akle, V. et al. F-spondin/sponlb expression patterns in developing and adult zebrafish. PLoS One 7, e37593 (2012).

[00448] 35. Hartman, B. H., Durruthy-Durruthy, R., Laske, R. D., Losorelli, S. & Heller, S. Identification and characterization of mouse otic sensory lineage genes. front. cell. neurosci. 9, 79 (2015).

[00449] 36. Petit, F., Sears, K. E. & Ahituv, N. Limb development: a paradigm of gene regulation. Nat. Rev. Genet. 18, 245-258 (2017).

[00450] 37. Guo, Q., Loomis, C. & Joyner, A. L. Fate map of mouse ventral limb ectoderm and the apical ectodermal ridge. dev. Biol. 264, 166-178 (2003).

[00451] 38. Lewandoski M, E. al. Fgf8 signaling from the AER is essential for normal limb development. - PubMed - NCBI. Available at: https://www.ncbi.nlm.nih.gov/pubmed/11101846. (Accessed: 22nd April 2018)

[00452] 39. Aoki M, E. al. R-spondin2 expression in the apical ectodermal ridge is essential for outgrowth and patterning in mouse limb development. - PubMed - NCBI. Available at: https://www.ncbi.nlm.nih.gov/pubmed/18067586. (Accessed: 22nd April 2018)

[00453] 40. Gerdes, J., Schwab, U., Lemke, H. & Stein, H. Production of a mouse monoclonal antibody reactive with a human nuclear antigen associated with cell proliferation. Int. J. Cancer 31, 13-20 (1983).

[00454] 41. Bergman, D., Halje, M., Nordin, M. & Engström W. Insulin-like growth factor 2 in development and disease: a mini-review. Gerontology 59, 240-249 (2013).

[00455] 42. Trapnell, C. et al. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat. Biotechnol. 32, 381-386 (2014).

[00456] 43. McInnes, L. & Healy, J. UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. (2018).

[00457] 44. Alexander Wolf, F. et al. Graph abstraction reconciles clustering with trajectory inference through a topology preserving map of single cells. bioRxiv 208819 (2017). doi:10.1101/208819

[00458] 45. Qiu, X. et al. Reversed graph embedding resolves complex single-cell trajectories. Nat. Methods 14, 979-982 (2017).

[00459] 46. Singh, M. K. et al. The T-box transcription factor Tbxl5 is required for skeletal development. Mech. dev. 122, 131-144 (2005).

[00460] 47. Paine-Saunders, S., Viviano, B. L., Zupicich, J., Skarnes, W. C. & Saunders, S. glypican-3 controls cellular responses to Bmp4 in limb patterning and skeletal development. dev. Biol. 225, 179-187 (2000).

[00461] 48. Hara, K. & Ide, H. Msx1 expressing mesoderm is important for the apical ectodermal ridge (AER)-signal transfer in chick limb development. dev. Growth Differ. 39, 705-714 (1997).

D. G. et al. Disruptions of Topological Chromatin Domains Cause Pathogenic Rewiring of Gene-Enhancer Interactions. Cell 161, 1012-1025 (2015).[00462] 49.

DG et al. Disruptions of Topological Chromatin Domains Cause Pathogenic Rewiring of Gene-Enhancer Interactions. Cell 161, 1012-1025 (2015).

[00463] 50. Davis, R. J. et al. Dachl mutant mice bear no gross abnormalities in eye, limb, and brain development and exhibit postnatal lethality. Mol. cell. Biol. 21, 1484-1490 (2001).

[00464] 51. Akiyama, H., Chaboissier, M.-C, Martin, J. F., Schedl, A. & de Crombrugghe, B. The transcription factor Sox9 has essential roles in successive steps of the chondr differentiationocyte pathway and is required for expression of Sox5 and Sox6. Gene Dev. 16, 2813-2828 (2002).

[00465] 52. Deng, Y. et al. Yap1 Regulates Multiple Steps of Chondrocyte Differentiation during Skeletal Development and Bone Repair. Cell Rep. 14, 2224-2237 (2016).

[00466] 53. Joshi, S. et al. TEAD transcription factors are required for normal primary myoblast differentiation in vitro and muscle regeneration in vivo. PLoS Genet. 13, e1006600 (2017).

[00467] 54. Knapp, D. et al. Comparative transcriptional profiling of the axolotl limb identifies a tripartite regeneration-specific gene program. PLoS One 8, e61352 (2013).

[00468] 55. Zeller, R., López-Rios J. & Zuniga, A. Vertebrate limb bud development: moving towards integrative analysis of organogenesis. Nat. Rev. Genet. 10, 845-858 (2009).

[00469] 56. Nishimoto, S., Minguillon, C, Wood, S. & Logan, M. P. 0. A combination of activation and repression by a colinear Hox code controls forelimb-restricted expression of Tbx5 and reveals Hox protein specificity. PLoS Genet. 10, el004245 (2014).

[00470] 57. Vargesson, N., Luria, V., Messina, I., Erskine, L. & Laufer, E. Expression patterns of Slit and Robo family members during vertebrate limb development. Mech. dev. 106, 175-180 (2001).

[00471] 58. Chimal-Monroy, J. et al. Analysis of the molecular cascade responsible for mesodermal limb chondrogenesis: Sox genes and BMP signaling. dev. Biol. 257, 292-301 (2003).

[00472] 59. Braun, T. & Gautel, M. Transcriptional mechanisms regulating skeletal muscle differentiation, growth and homeostasis. Nat. Rev. Mol. Cell biol. 12, 349-361 (2011).

[00473] 60. Tajbakhsh, S., Rocancourt, D., Cossu, G. & Buckingham, M. Redefining the genetic hierarchies controlling skeletal myogenesis: Pax-3 and Myf-5 act upstream of MyoD. Cell 89, 127-138 (1997).

[00474] 61. Harel, I. et al. Distinct origins and genetic programs of head muscle satellite cells. dev. Cell 16, 822-832 (2009).

[00475] 62. Sambasivan, R. et al. Distinct regulatory cascades govern extraocular and pharyngeal arch muscle progenitor cell fates. dev. Cell 16, 810-821 (2009).

[00476] 63. Heimberg, G., Bhatnagar, R., El-Samad, H. & Thomson, M. Low Dimensionality in Gene Expression Data Enables the Accurate Extraction of Transcriptional Programs from Shallow Sequencing. Cell Syst 2, 239-250 (2016).

[00477] 64. Cusanovich, D. A. et al. The cis-regulatory dynamics of embryonic development at single cell resolution. (2017). doi:10.1101/166066

[00478] 65. McKenna, A. et al. Whole-organism lineage tracing by combinatorial and cumulative genome editing. Science 353, aaf7907 (2016).

[00479] 66. Osterwalder, M. et al. Enhancer redundancy provides phenotypic robustness in mammalian development. Nature 554, 239-243 (2018).

[00480] 67. Dickel, D. E. et al. Ultraconserved Enhancers Are Required for Normal Development. Cell 172, 491-499.el5 (2018).

[00481] 68. Li, D. et al. Formation of proximal and anterior limb skeleton requires early function of Irx3 and Irx5 and is negatively regulated by Shh signaling. dev. Cell 29, 233-240 (2014).

[00482] 69. Kraft, K. et al. Deletions, Inversions, Duplications: Engineering of Structural Variants using CRISPR/Cas in Mice. Cell Rep. (2015). doi:10.1016/j.celrep.2 015.01.016

[00483] 70. Buenrostro, J. D., Giresi, P. G., Zaba, L. C., Chang, H. Y. & Greenleaf, W. J. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat. Methods 10, 1213-1218 (2013).

[00484] 71. Renaud, G., Stenzel, U., Maricic, T., Wiebe, V. & Kelso, J. deML: robust demultiplexing of Illumina sequences using a likelihood-based approach. Bioinformatics 31, 770-772 (2015).

[00485] 72. Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15-21 (2013).

[00486] 73. Anders, S., Pyl, P. T. & Huber, W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics btu638 (2014).

[00487] 74. Qiu, X. et al. Reversed graph embedding resolves complex single-cell developmental trajectories. (2017). doi:10.1101/110668

[00488] 75. Wolf, F. A., Angerer, P. & Theis, F. J. SCANPY: large-scale single-cell gene expression data analysis. Genome Biol. 19, 15 (2018).

[00489] 76. Cao, J. et al. Comprehensive single-cell transcriptional profiling of a multicellular organism. Science 357, 661-667 (2017).

[00490] 77. Pliner, H. et al. Chromatin accessibility dynamics of myogenesis at single cell resolution. (2017). doi:10.1101/155473

[00491] 78. Kuleshov, M. V. et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 44, W90-7 (2016).

[00492] 79. McInnes, L. & Healy, J. UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. (2018).

[00493] 80. Levine, J. H. et al. Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with Prognosis. Cell 162, 184-197 (2015).

[00494] 81. Wolf, F. A. et al. Graph abstraction reconciles clustering with trajectory inference through a topology preserving map of single cells. (2017). doi:10.1101/208819

[00495] 82. Wagner, D. E. et al. Single-cell mapping of gene expression landscapes and lineage in the zebrafish embryo. science ear4362 (2018).

[00496] 83. Briggs, J. A. et al. The dynamics of gene expression in vertebrate embryogenesis at single-cell resolution. science ear5780 (2018).

[00497] 84. Mao, Q., Wang, L., Tsang, I. & Sun, Y. Principal Graph and Structure Learning Based on Reversed Graph Embedding. IEEE Trans. pattern anal. Mach. Intel. (2016). doi:10.1109/TPAMI.2 016.2 635657

[00498] 85. Qiu, X. et al. Reversed graph embedding resolves complex single-cell trajectories. Nat. Methods 14, 979-982 (2017).

[00499] 86. Moran, P. A. P. Notes on continuous stochastic phenomena. Biometrika 37, 17-23 (1950).

[00500] Example 2

[00501] A new method for isolating nuclei from tissue and fixing them (sc-RNA-seq)

[00502] Reagents: BSA (molecular biology grade, NEB, #B9000S); Super-RNase inhibitor (Thermo, No. AM2696); EMS 157-4-100, 4% paraformaldehyde (formaldehyde) aqueous solution, EM grade, 100 ml (Amazon).

[00503] Buffers: Buffer for nuclei (stored at 4°C): 10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl ₂ , 10% IGEPAL CA-630 (stored at 4°C). Nuclear wash buffer (replaced fresh each time): 980 µl of nuclear buffer with 10 µl of BSA and 10 µl of Super-RNase inhibitor were thoroughly mixed and stored on ice. Nuclear lysis buffer (replaced fresh each time): Nuclear wash buffer with 0.1% IGEPAL CA-630.

[00504] Isolation of nuclei directly from tissue

[00505] Tissues were minced into small pieces with a razor blade in 1 ml ice-cold cell lysis buffer (10 mM Tris, pH 7.4, 10 mM NaCl, 3 mM and 0.1% IGEPAL CA-630, 1% inhibitor Super -RNase and 1% BSA) and transferred to the top of a 40 µm cell filter (Falcon).

[00506] Tissues were homogenized with a rubber tip syringe (5 ml, BD) in 4 ml cell lysis buffer.

[00507] Then, the filtered nuclei were transferred to a new 15 ml tube (Falcon) and pelleted by centrifugation at 500×g for 5 minutes and washed once with 1 ml of cell lysis buffer.

[00508] Fixing nuclei

[00509] The nuclei were fixed in 4 ml of ice-cold 4% paraformaldehyde (EMS) for 15 minutes on ice.

[00510] After fixation, nuclei were washed twice in 1 ml of nuclear wash buffer (cell lysis buffer without IGEPAL) and resuspended in 500 μl of nuclear wash buffer.

[00511] Samples were divided into several batches and flash frozen in liquid nitrogen. Frozen samples may be transferred to dry ice.

[00512] Example 3

[00513] Characterization of the dynamics of cell transition from one state to another by sci-fate

[00514] The development of a living organism consists in the passage of cells of various states in a strictly organized temporal order. Despite the increasing use of genomic methods for individual cells, the quantitative assessment of the dynamics of the transition of a cell from one state to another remains a difficult task. In this case, the authors introduce the concept of "sci-fate" as a high-throughput analysis based on combinatorial indexing, carried out to profile both the whole and the newly synthesized transcriptome in each of the thousands of individual cells. As proof of this concept, the authors applied the concept of sci-fate to a model cortisol response system and characterized more than 6000 cell transition events from one state to another according to the known dynamics of cell cycles after glucocorticoid receptor activation. Based on this analysis, the authors showed that the direction of the cell transition from one state to another and its probability are regulated by the distances between these states and the nature of the instability of such states. This technique and computational methods can be easily applied to other biological systems to quantitatively characterize the dynamics of cell states and decipher the internal mechanism for determining cell fate.

[00515] A cell goes through various functional and molecular states during the development of a multicellular organism. Characterization of a cell's transition pathway from one state to another, or cell fate, is the basis for understanding developmental processes and their applications in fields such as cell engineering. Despite the increasing use of genomic methods for single cells, these methods only provide images of the state of the cell and therefore cannot provide information on the dynamics of cell transitions (1). Although single-cell tracking using time-of-flight microscopy can be used to characterize cellular transitions from one state to another (2, 3), these methods have limited throughput and can track changes in only a few genes, and therefore have a low capacity. to decipher complex systems.

[00516] In this application, the authors have described a new strategy for quantifying the dynamics of the transition of a cell from one state to another at the level of the entire transcriptome. This strategy relies on a new RNA-seq technique for single cells based on combinatorial indexing, sci-fate. By labeling newly synthesized mRNA with 4-thiouridine (4, 5), which will give rise to C>T point mutations during reverse transcription, sci-fate captures both the entire transcriptome and the newly synthesized transcriptome at the individual cell level, along with information about the degraded transcriptome in the past (memory of the previous state). The memory of the previous states of each cell is then corrected for the rate of mRNA degradation (memory correction method), so that each cell can be characterized by the dynamics of the transcriptome between two time points.

[00517] To characterize the dynamics of cell transition from one state to another, regulated by internal and external factors, the authors applied the concept of sci-fate to a model system of response to cortisol, where cell fate is triggered by two main factors: the program of the internal cell cycle and external activation of the glucocorticoid receptor (GR) induced by the drug. GR activation affects the activity of almost every cell in the body and regulates genes that control development, metabolism, and immune response (6). Using sci-fate, the authors determined the dynamics of the entire transcriptome for more than 6000 individual cells. Based on the similarity between the past and current states of the transcriptome, the authors constructed thousands of trajectories for the transition of cells from one state to another, covering five time points, which can be grouped into three types of cell fate in accordance with the known patterns of the cell cycle during GR activation. In addition, the authors characterized the latent cellular states by the functional activity of TF modules, and concluded a network of cellular transitions to predict the state of the cell. Finally, the authors showed that the direction and probability of cell transition from one state to another are regulated by the similarity of transcriptomes and the nature of the instability of neighboring states. The theoretical, computational, and experimental approaches developed here should be easily applicable to other biological systems in which the dynamics of cellular transitions is still unknown.

[00518] Overview of sci destiny

[00519] The sci fate analysis is carried out by carrying out the following steps (FIG. 30A): (i) cells are first incubated with 4-thiouridine (S4U), a commonly used thymidine analog, to label the newly synthesized RNA (7-13). (ii) Cells are harvested, fixed with 4% paraformaldehyde, followed by an alkylation reaction by coupling with thiol (SH), which results in the covalent attachment of a carboxyamide methyl group to S4U via nucleophilic substitution (4). (iii) Cells are distributed by weight into each well of 4 x 96 well plates. The first RNA-seq molecular index is introduced into the cell mRNA in each well by in situ reverse transcription (RT) using a poly(T) primer having a well-specific barcode and a degenerate unique molecular identifier (UMI). During cDNA synthesis, mRNA labeled with modified S4U mimics the conversion of thymine to cytosine (T>C) and leads to the formation of a mutated first strand cDNA. (iv) Cells from all wells are pooled and then redistributed by fluorescence activated cell sorting (FACS) in multiple 96-well plates. Cells are gated by DAPI (4',6-diamidino-2-phenylindole) staining to distinguish single cells from doublets during sorting. Double-stranded cDNA is prepared by RNA degradation and second strand synthesis and transposed with Tn5. The cDNA is then amplified by polymerase chain reaction (PCR) with a combination of primers recognizing the α-adapter at the 5' end and the RT primer at the 3' end. These primers also have a well-specific barcode that introduces a second RNA-seq molecular index. (v) PCR amplicons are pooled and subjected to massive parallel sequencing. As with other "sci" protocols (14-21), most nuclei pass through a unique combination of wells, and therefore the contents of each well are labeled with a unique combination of barcodes that can be used to group reads originating from the same same cells. Newly synthesized mRNA from the entire transcriptome is identified by T>C transformations, corrected for background error (Method).

[00520] To evaluate quality control, we first tested the method using a mixture of cells (human) and NIH/3T3 (mouse) under four conditions: S4U labeled or unlabeled (200 nM, 6 hours) and with or without IAA treatment (FIGS. 31A-D). When labeled with S4U and treated with IAA (sci fate state), transcriptomes from human/mouse cells were highly species-specific (>99% purity for human and mouse cells, 2.6% collisions) with a high ratio of mutated T reads detected. >C (46% for human cells and 31% for mouse cells in the sci-fate state compared to 0.8% for human cells and 0.8% for mouse cells under untreated conditions). Nearly equivalent cell purity was achieved by the authors under the four conditions, although slightly lower UMIs were found in the IAA treatment groups. Pooled transcriptomes after sci-fate versus normal sci-RNA-seq were highly correlated (Spearman correlation r=0.99; Figures 31E-F), suggesting that the short-term labeling and conversion process has minimal effect on the state of the cell.

[00521] Joint profiling of the total and newly synthesized transcriptome in A549 cells treated with dexamethasone.

[00522] The authors then applied sci-fate analysis to a cortisol response model where dexamethasone (DEX), a synthetic cortisol mimic, activates the glucocorticoid receptor (GR), which binds to thousands of positions throughout the genome and significantly alters cell state in a short amount of time ( 22-25). We treated A549 cells derived from lung adenocarcinoma for 0, 2, 4, 6, 8, or 10 hours with 100 mM DEX. In each condition, cells were incubated with S4ET (200 nM) for the last two hours before cell collection in 384 x 192 wells for sci-fate assessment (Fig. 30B). Each of these six conditions is presented in 64 wells during the first indexing cycle so that the processing conditions can be reconstructed based on the first index of each cell.

[00523] After filtering out low-quality cells, potential doublets, and a small subset of differentiated cells (Method), single cell profiles were determined for 6680 cells (median for 26176 mRNA detected, per cell) with a median of 20% labeled UMI per cell (Fig. 30C , Fig. 32A-B). Intron reads showed a significantly higher level of newly synthesized mRNA than exon reads (65% in intron reads and 13% in exon reads, p-value <2.2e-16, Wilcoxon's sign-rank test; Fig. 30D), as expected. that intron reads will be enriched in the newly synthesized transcriptome.

[00524] First, the authors wondered whether the entire transcriptome and the newly synthesized transcriptome convey different information on the characterization of the state of the cell. The authors pooled the entire transcriptome and the newly synthesized transcriptome for each processing condition and assessed their correlation. In contrast to the entire transcriptome, the newly synthesized transcriptome showed a sharp difference between the non-DEX-treated (0 h) and DEX-treated groups (FIG. 32C). Consistently, dimensionality reduction by uniform multiple approximation and projection (UMAP) (26) for the entire or newly synthesized transcriptome gives different results (Fig. 30E): the entire transcriptome cannot separate cells that have not been treated with DEX (0 h .), and cells with early DEX treatment (2 hours), while the newly synthesized transcriptome combines all DEX-treated cells into one group. Cell clusters identified from the whole or newly synthesized transcriptome do not completely match each other (Fig. 30F, Fig. 32D-E). This was expected, since the newly synthesized transcriptome directly reflects the activity of the gene promoter or the epigenetic response to the external environment, while the entire transcriptome is mainly determined by the mRNA residue from its previous state.

[00525] To characterize the condition of cells using general information, the authors combined the principal components (PC) from the whole and newly synthesized transcriptome for UMAP analysis. General information divided cells into cells without DEX-treatment (0 hours), cells with early treatment (2 hours) and late treatment (>2 hours) (Fig. 30E). Interestingly, two clusters (cluster 1 and 4) characterized by the whole transcriptome were divided into four separate groups by common information (Fig. 30F). The authors assessed the level of expression and the rate of new synthesis of gene markers associated with the cell cycle (27) (Fig. 30G, Fig. 32F-G): according to the general information, the newly separated clusters correspond to cells in the G2/M phase (high expression level and high rate synthesis of G2/M markers) and cells in the early G0/G1 phase (high expression level and low rate of synthesis of G2/M markers). This suggests that the newly synthesized transcriptome conveys different information about the state of the cell compared to the entire transcriptome, and general information can provide higher resolution when characterizing the state of the cell.

[00526] Characterization of TF functional modules to determine cell fate.

[00527] Further, the authors have made an attempt

to characterize the TF modules that trigger the process of cell transition from one state to another. Associations between transcription factors (TF) and their regulated genes were identified in two steps: for each gene, the authors calculated the correlations between the rate of mRNA synthesis during the last two hours and the level of TF expression for more than 6000 cells using LASSO (least absolute shrinkage and operator selection). These identified associations were further filtered either by analysis of published CHIP-seq data (28) or analysis for motif enrichments (29) (Method). A total of 986 links between 29 TF and 532 genes were identified (FIG. 33A, Table S1) based on TF gene covariance and validation of DNA binding data. To assess the possibility that associations are artifacts of controlled regression, the authors permuted the TF expression matrix sample identifiers and performed the same analysis. After this rearrangement, no links were found.

[00528] TF modules that trigger a GR response have been identified, including known GR response effectors such as CEBPB (30) (FIGS. 34A-B), FOXOl (37), and JUNB (32) (FIGS. 33A). The authors also found several new TF modules associated with the GR response, including YOD1 and GTF2IRD1, with increased expression and activity in DEX-treated cells (Fig. 34C-D). Major TF modules that trigger the cell cycle process have been identified and these modules include E2F1, E2F2, E2F7, BRCA1 and MYBL2 (33). Compared to the overall level of expression, the rate of synthesis of novel RNA genes regulated by cell cycle TF modules correlated more strongly with the expression of the TF target (Fig. 34E). In addition, the authors also found TF modules associated with cell differentiation, such as GATA3, mainly expressed in the resting cell population group (34), and TF modules associated with oxidative stress response, such as NRF1 (35) and NFE2L2 (NRF2) (36).

[00529] We then characterized TF activity by combining the rates of new RNA gene synthesis in each TF module and calculated the absolute correlation coefficient between each pair of TFs (FIG. 34F). The highly correlated TF activity indicates that these pairs may function in combination with each other. Hierarchical clustering divides these 29 TF modules into five main modules (Fig. 34F): the first module represents all cell cycle associated TF modules such as E2F1 and FOXM1 (33) and is the trigger for the cell cycle process. The third module is all TF modules associated with the GR response such as FOXOl, CEBPB, JUNB and RARB (30) (31) (32). Other groups of TF modules include three TFs (KLF6, TEAD1 and YOD1) co-regulated by both the cell cycle and the GR response (module 2); internal differentiation pathway, including GATA3 and AR (module 3), and TFs associated with stress response such as NRF1 and NFE2L2 (module 5).

[00530] To identify different cell cycle states, we first ordered cells by activity of the TF module associated with the cell cycle. Cells were ordered along a smoothed cell cycle trajectory, as evidenced by the rate of synthesis of known cell cycle markers (27) (Fig. 33B). The authors observed a gap between the G2/M phase and the G0/G1 phase, which corresponds to a sharp change in the state of the cell during its division. Through uncontrolled clustering, the authors identified nine cell cycle states spanning G0/G1, S, and G2/M cell cycle phases based on cell cycle marker expression level (FIG. 33B). Cells can be ordered along a different smoothed trajectory using the TF modules associated with the GR response. This trajectory clearly correlates with DEX treatment time and the dynamics of known TF activity regulated by GR activation (FIG. 33C). Through uncontrolled clustering analysis, the authors identified three cell clusters by GR responses, corresponding to no GR response/low/high GR response (FIG. 33C).

[00531] We then attempted to quantify the latent cellular states in the system (FIG. 35A). Nine cell cycle states and three GR response states were shown in FIG. 33B-S. All possible combinatorial conditions were identified, with the smallest group comprising 1.1% (74) of all cells (FIG. 33D). The observed proportion of cellular states was close to that expected under the condition of independent selection. This corresponds to a low correlation coefficient (Pearson correlation r=0.004) between the activity of these two functional TF modules for more than 6000 cells. For comparison, by dimensionality reduction and clustering analysis for the entire and newly synthesized transcriptome, 6 major clusters were identified by the authors (FIG. 35B). These major clusters can be easily identified by pooling the groups of these 27 cell states (FIG. 33E).

[00532] Characterization of the trajectory of the transition of cells from one state to another and the network of transition of such states

[00533] Using the entire transcriptome and the newly synthesized transcriptome characterized for each cell, the authors were able to infer the state of the transcriptome of one cell prior to labeling with S4U (FIG. 36A). Restoration of the previous cell transcriptome depends on two parameters: the rate of detection of newly synthesized reads in terms of sci-fate and the rate of degradation (or semi-degradation) of each mRNA (Method). Both of these parameters can be evaluated in the same sci-fate experiment.

[00534] First, the authors evaluated the level of detection of sci fate. The authors suggest that the half-life of mRNA is constant under various conditions of DEX-treatment. This assumption was also confirmed by a subsequent stability test. According to this assumption, the total transcriptome partially disrupted 2 hours prior to S4U labeling should be the same in non-DEX-treated cells and in DEX-treated cells 2 hours later. Thus, the differences in the total transcriptome (volume) should match the differences in the newly synthesized transcriptome (volume), adjusted for the degree of detection depending on the technique. Since the entire transcriptome and the newly synthesized transcriptome were characterized in this experiment, the authors could directly determine the level of sci fate detection. Differences in newly synthesized mRNA correlate well with differences in mRNA expression level (Pearson's test, r=0.93, Fig. 37A), suggesting that the level of detection of new RNA is fairly stable across all genes. Thus, the authors used the median novel RNA uptake rate (82%) for subsequent analysis.

[00535] The degree of mRNA degradation was then calculated for 2 hours. Since the A549 cell population can be considered stable without external perturbation, for cells after 2 hours of DEX treatment, their previous state (2 hours before labeling with S4U) should be the same as for cells not treated with DEX. Similarly, the previous state (before labeling with S4U) for cells treated with DEX at time T=0/2/4/6/8/10 hours should be similar to the state of profiled cells at time T=0/0/2/4/6 /8 ocloc'k. With respect to the whole transcriptome and the newly synthesized transcriptome characterized for all processing conditions, it is possible to estimate the degree of mRNA degradation for several thousand genes in each 2-hour time interval. After the aforementioned stability test, it was found that the extent of gene disruption correlated significantly with DEX treatment time (FIG. 37B). The authors then used the average degree of gene disruption for further analysis. After determining the average degree of detection of new mRNA and the degree of gene destruction, the authors evaluated the previous state of the transcriptome of one cell so that each cell could be characterized by the dynamics of the transcriptome in a two-hour interval.

[00536] To assess the dynamics of cell states over a longer interval (i.e., within 10 hours), the authors developed a streaming algorithm to evaluate the relationship of parent and daughter cells in the same trajectory of transition of cells from one state to another (Fig. 36A): For each A cell (eg, DEX-treated cells at 2 hours), the profile of cell B at an earlier time point was identified (eg, for cells not treated with DEX), and the current state of cell B was similar to its previous state A, as determined from a recently developed alignment strategy to identify common cell states between two datasets (27). State B can be considered as the parent state of A. Similarly, the authors also determined the profile of another cell C at a later time point (for example, for cells treated with DEX after 4 hours), and cell C had its previous state similar to the current state of cell A Cell C can be considered as the future state of cell A. Applying the same strategy to all previous and subsequent states defined for each cell, the authors constructed 6680 cell transition trajectories from one state to another within 10 hours and at five time points (Fig. 36A-B). It should be noted that this analysis is based on the assumption that the previous and current states of each cell (except for cells at the initial and final time points) were detected in detail, which is also true for the data sets obtained by the authors, since more than 6000 cells were profiled (more than 1000 cells per condition), or one cell for less than one minute during the cell cycle. The state profile of each of the many cells (>50) was determined, and thus the stochastic process of cell transition from one state to another can also be recorded.

[00537] To confirm the result, the authors performed a dimensionality reduction and uncontrolled clustering analysis for the trajectories of 6680 individual cells, which were grouped into three clusters of trajectories. The authors evaluated the dynamics of the cellular states characterized in Figs. 36C. As expected, all three trajectories showed a transition of cells from one state to another in the absence of a GR response at a low/high GR response for a certain period of time (Fig. 36D). The authors observed different dynamics of the cell cycle in all these three trajectories (Fig. 36D): trajectory 1 indicated a decrease in the G2/M phase and, accordingly, an increase in the G0/G1 phase, and a transition of the cell from the intermediate states G2/M and G1 to the phase G1. Trajectory 2 indicated the transition of the cell from the intermediate states S and G2/M to the G2/M phase. In trajectory 3, the authors observed a cell transition from intermediate G1 and S phase to early S phase during early DEX treatment (0-2 hours), but this transition was inhibited under conditions of later DEX treatment (>2 hours after DEX treatment), indicating that long-term DEX treatment causes the G1 phase to stop. This is consistent with changes in the ratios of cell states over a certain treatment time and with data from previous studies (37,38) (Fig. 36D). This suggests that the transition pathways of individual cells, characterized by sci-fate, can restore the general directions of cell transition from one state to another.

[00538] After determining the profile of each state of a plurality of cells (>70), the authors determined the probability of transition of cells from one state to another in all 27 latent states. Cell transitions from one state to another with a low transition probability (<0.1) can be caused by rare events or noise and thus filtered out. The network of cell transitions from one state to another can be defined by the 2 7 cell states as nodes, and the links indicate possible transition paths (FIG. 36E). The direction of passage of the cell cycle can be easily characterized by at least three stages of transition with irreversible directions of transition during the cell cycle (Fig. 36E). In the late G1 phase and late G2/M phase, the authors also found several conditions indicating reversible transitional dynamics that may indicate two cell cycle checkpoints in the G1/S and G2/M phases (33). As expected, cells with the same cell cycle but different GR responses showed dramatically different transition dynamics, and cells with a high GR response generally tend to G1 or G2/M phase arrest.

[00539] As a consistency check, in order to confirm whether the cell transition network from one state to another can capture the dynamics of cell transition from one state to another, the authors conducted an analysis to determine whether the transition probabilities can indicate the possibility of restoring the real distributions of cell states at different points in time. Indeed, although cell state ratios change dynamically over 10 hours (FIG. 36F), however, the state transition network accurately predicts the ratios of the 27 cell states at all five later time points, starting from the state ratio of DEX-treated cells at 0 hours ( Fig. 36G, Fig. 38A). The authors also computed a cell state transition network for only part of the data (from 0 to 6 hours), which gave highly correlated transition probabilities with the full data, and accurately predicted cell states 10 hours in advance (FIG. 36H, FIG. 38B).

[00540] Characterization of factors regulating the direction of cell transition from one state to another

[00541] To characterize the factors governing the likelihood of cell state transition, we first calculated the distance between cell states from the Pearson distance for the combined transcriptome (whole and newly synthesized) between each pair of states. As expected, the probability of a cell transition from one state to another is negatively correlated with transition distance (Spearman correlation coefficient = -0.38, Fig. 39A). The authors also calculated the instability of the state, defined by the proportion of cells leaving this state within two hours (Fig. 39B). The nature of state instability correlates well with cell transition directions (FIG. 39B): states in the absence of a GR response are more unstable compared to states with a high GR response. In states with a high GR response, cells in early G1 phase have the lowest instability, while cells in intermediate G1/S states show a high peak of instability, consistent with G1 phase arrest with later DEX treatment.

[00542] Changes in cell state ratio after 10 hours correlate well with cell state instability (Spearman correlation coefficient = -0.88, Fig. 39C), suggesting that cell state dynamics depend on the nature of cell state instability. State instability also correlates well with state transition probability entropy, indicating a variety of state transition endpoints (Pearson correlation r=0.73, FIG. 39D). In order to test whether the probability of transition between states can be a consequence of the instability of neighboring states, the authors plotted the dependence of instability on the distance to the nearest states in a neural network model to predict the probability of transition from each state to other states. The combination of the instability of neighboring states and distances made it possible to increase the efficiency of predicting the probability of transition between one state to another more than ten times compared to using only distances between states (after cross-validation, the median criterion r ² is 0.58 when using information about two parameters , and 0.046 using only distances between states, p-value = 4.5e-10, two tailed Wilcoxon rank sum test, Fig. 39E), suggesting that the directions and probabilities of cell state transition depend on the nature of neighboring state stability. At the same time, cells prefer to move to a more stable immediately next neighboring state.

[00543] Discussion

[00544] The authors of the present application developed the first strategy for characterizing the dynamics of the transition of the state of cells at the level of the entire transcriptome. This strategy is sci-fate dependent, i.e., it is a new high-throughput method for sequencing the RNA sequence of individual cells based on combinatorial indexing, which makes it possible to characterize both the entire and newly synthesized transcriptome in thousands of cells. Like other sci methods, sci fate can be easily scaled to millions of cells (39) and is potentially compatible with both transcriptome and epigenome characterization (40). This sci-fate can be used to characterize the dynamics of the state of cells in a much more complex system (that is, at the level of development of the entire embryo), where the real path of cell transition into cells of several hundred types is still unknown. In addition, the authors developed a flow-type computational algorithm to assess the degree of capture of newly synthesized RNA and the degree of gene destruction according to sci-fate (memory correction), and thousands of differential trajectories were built for each of the individual cells connected by a common previous and current state. transcriptome at any given time.

[00545] To test the methods and to determine how intrinsic and extrinsic factors influence the dynamics of cell states, the authors applied this strategy to a cortisol response model system where cell fate is dynamically regulated by inner cell cycle GR activation and extrinsic GR activation, caused by the drug. The authors showed that the newly synthesized transcriptome is directly related to the response of the epigenome to external stimuli, and the joint analysis of the whole and newly synthesized transcriptome revealed higher resolution in the separation of cellular states. By covariating between TF expression and the rate of new RNA synthesis in thousands of cells, the authors identified up to a thousand links between TF and regulated genes as indicated by DNA binding data. The authors also identified 27 "hidden cellular states" characterized by the combinatorial state of TF functional modules in cell cycle progression and GR response, compared to only 6 states assessed by standard clustering analysis.

[00546] By analyzing for memory correction and cellular connections, the authors constructed more than 6000 individual cell transition trajectories spanning 10 hours, with the main trajectories corresponding to the known dynamics of the state of the cell in the cell cycle and the GR response. The network of cell transition from one state to another is characterized by the probability of transition in all cell states, which is confirmed by the restoration of the dynamics of 27 cell states at all five time points. Finally, the authors found that the probabilities of a cell transition from one state to another depend on two key factors of the cell state transition network: the distance between states and the nature of state instability, both of which can be estimated by conventional -seq sequencing methods for single cells.

[00547] This strategy, while powerful, has several limitations. First, to accurately construct the trajectory of an individual cell, a complete characterization of the state of the cell at each moment of time is necessary. To accurately estimate the transition probability, it is also necessary to make many observations for each state. These limitations can be easily overcome using the sci-fate combinatorial strategy, which allows the characterization of millions of cells in a single experiment. Another limitation is that in vitro systems require many S4U labeling experiments. However, recent studies have shown that S4U allows for stable labeling of a particular cell type through specific RNA transcription in a variety of mouse tissues (i.e., brain, gut, and adipose tissue) (41, 42), suggesting that sci-fate estimation, together with further optimizations to increase the level of S4U incorporation and the degree of detection, can be applied to profile the dynamics of the transcriptome of individual cells in vivo.

[00548] The assessment of sci-fate opens up new possibilities for applying "static" genomic methods to characterize the dynamic systems of individual cells. Compared to traditional imaging-based technologies, sci-fate assessment allows characterization of cell state dynamics at the level of the entire transcriptome and provides a comprehensive characterization of cell state without selecting markers and revealing the key driving force in cell differentiation. Finally, the authors suggest that sci-fate can be easily combined with alternative lineage tracing methods (43-45) to decipher the detailed dynamics of cell state transition to each final state within a few hundred lineages.

[00549] Materials and methods.

[00550] Mammalian cell culture.

[00551] All mammalian cells were cultured at 37° C. with 5% CO ₂ and stored in high glucose DMEM (Gibco, Cat. No. 11965) for HEK293T and NIH/3T3 cells, or in DMEM/F12 medium for A549 cells, to which was added 10% FBS and 1x Pen/Strep (Gibco, cat. no. 15140122; 100 U/ml penicillin, 100 μg/ml streptomycin). Cells were trypsinized with 0.25% trypsin-EDTA (Gibco, cat. no. 25200-056) and distributed at a ratio of 1:10 three times a week.

[00552] Processing samples for sci destiny

[00553] A549 cells were treated with 100 nM DEX for 0 hours, 2 hours, 4 hours, 6 hours, 8 hours and 10 hours. Cells under all treatment conditions were incubated with 200 μM S4U for the last two hours before harvesting the cells. For HEK2 93T and NIH/3T3 cells, cells were incubated with 200 μM S4U for 6 hours before harvesting the cells.

[00554] All cell lines (A549, HEK293T and NIH/3T3 cells) were trypsinized, centrifuged at 300xg for 5 minutes (4°C) and washed once in ice-cold PBS. All cells were fixed with 4 ml ice-cold 4% paraformaldehyde (EMS) for 15 minutes on ice. After fixation, cells were pelleted at 500×g for 3 minutes (4°C) and washed once with 1 ml PBSR (1× PBS, pH 7.4, 1% BSA, 1% SuperRnaseln, 1% 10 mM DTT). After washing, the cells were resuspended in PBSR at 10 million cells per ml, flash frozen and stored in liquid nitrogen. Cells fixed with paraformaldehyde were thawed in a water bath at 37° C., centrifuged at 500×g for 5 minutes and incubated with 500 μl of PBSR containing 0.2% Triton X-100 for 3 minutes on ice. Cells were pelleted and resuspended in 500 µl of nuclease-free water containing 1% SuperRnaseln. Then, 3 ml of 0.1 N was added to the cells and incubated for 5 minutes on ice (21). 3.5 ml of Tris-HCl (pH=8.0) and 35 μl of 10% Triton X-100 were added to the cells for neutralization. Cells were pelleted and washed with 1 ml PBSR. Cells were resuspended in 100 µl PBSR. 100 µl PBSR with fixed cells were incubated with a mixture containing 40 µl iodoacetamide (IAA, 100 mM), 40 µl sodium phosphate buffer (500 mM, pH=8.0), 200 µl DMSO and 20 µl H ₂ O, at 50 °C for 15 minutes. The reaction was quenched with 8 μl DTT (1 M) and 8.5 ml PBS (47). Cells were pelleted and resuspended in 100 μl PBSI (1×PBS, pH 7.4, 1% BSA, 1% SuperRnaseln). For all subsequent washes, nuclei were pelleted by centrifugation at 500×g for 5 minutes (4°C).

[00555] The following steps were performed according to the sci-RNA-seq protocol for paraformaldehyde-fixed nuclei (15, 16). Briefly, cells were distributed into four 9b-well plates. For each well, 5000 nuclei (2 µl) were mixed with 1 µl of 25 mM anchored oligo-dT primer (5'-ACGACGCTCTTCCGATCTNNNNNNNN[10 bp index]

TTTTTTTTTTTTTTTTTTTTTTTTTTTTVN-3') (SEQ ID NO: 5), where "N" means any base and "V" means "A", "C" or "G"; IDT), and 0.25 μl of a mixture of 10 mm dNTP (Thermo) were denatured at 55°C for 5 minutes and immediately placed on ice. Thereafter, 1.75 µl of the first strand reaction mixture containing 1 µl of 5× Superscript IV first strand buffer (Invitrogen), 0.25 µl of 100 mM DTT (Invitrogen), 0.25 µl of Superscript reverse transcriptase was added to each well. IV (200 U/µl, Invitrogen), 0.25 µl of recombinant RNaseOUT ribonuclease inhibitor (Invitrogen). Reverse transcription was performed by incubating the plates in a temperature gradient: 4°C, 2 minutes, 10°C, 2 minutes, 20°C, 2 minutes, 30°C, 2 minutes, 40°C, 2 minutes, 50°C, 2 minutes, and 55°C, 10 minutes. Then, all cells (or nuclei) were pooled, stained with 4',6-diamidino-2-phenylindole (DAPI, Invitrogen) at a final concentration of 3 μM, and sorted at 25 nuclei per well in 5 μl of EB buffer. Cells were gated by DAPI staining so that singlets were separated from doublets and sorted into each well. Then, 0.66 µl of second strand mRNA synthesis buffer (NEB) and 0.34 µl of second strand mRNA synthesis enzyme (NEB) were added to each well, and second strand synthesis was performed at 16° C. for 180 minutes. The contents of each well were then mixed with 5 µl of Nextera TD buffer (Illumina) and 1 µl of TDE1 enzyme for i7 only (25 nM, Illumina diluted in Nextera TD buffer) and incubated at 55°C for 5 minutes for labeling. The reaction was terminated by adding 10 μl of DNA binding buffer (Zymo) and incubating at room temperature for 5 minutes. Each well was then purified using 30 µl AMPure XP spheres (Beckman Coulter), eluted in 16 µl EB buffer (Qiagen) and then transferred to a fresh multi-well plate.

[00556] For the PCR reaction, the contents of each well were mixed with 2 µl 10 µM primer P5 (5'-AATGATACGGCGACCGAGAATCTACAC[i5]ACACTCTTTTCCC TACACGACGCTCTTCCGAT CT-3'; IDT) (SEQ ID NO: 6), 2 µl 10 µM primer P7 (5'-CAAGCAGAAGACGGCATACGAGAT[i7]GTCTCGTGGGCTCGG-3', IDT) (SEQ ID NO: 7) and 20 µl 2× NEBNext High-Fidelity Master Mix (NEB) PCR mix. Amplification was carried out according to the following program: 72°C for 5 minutes, 98°C for 30 seconds, 18-22 cycles (98°C for 10 seconds, 66°C for 30 seconds, 72°C for 1 minute ) and finally at 72°C for 5 minutes. After PCR, samples were pooled and purified using 0.8 volume of AMPure XP spheres. Library concentrations were determined with Qubit (Invitrogen) and libraries visualized by electrophoresis on a 6% TBE-PAGE gel. All libraries were sequenced on a single NovaSeq 500 platform (Illumina) using the V2 150 cycle kit (Reid 1: 18 cycles, Reid 2: 130 cycles, Index 1: 10 cycles, Index 2: 10 cycles).

[00557] Read alignment and post-processing

[00558] Read alignment and generation of a gene count matrix for single cell RNA-seq sequencing was performed using a stream-type algorithm that was developed for sci-RNA-seq (48), with minor modifications. First, reads were mapped to a reference genome using STAR/v2.5.2b {49) with gene annotations from GENCODE V19 for human and GENCODE VM11 for mouse. For experiments with HEK2 93T and NIH/3T3 cells, an index was used that included a combination of human (hgl9) and mouse (mm10) chromosomes. For the A549 experiment, the authors used the hgl9 human genome construct.

[00559] The individual cell sam files were first converted to an alignment tsv file using jvarkit's sam2tsv function (50). Then, to create an alignment file for each individual cell, mutations corresponding to background SNPs were filtered out. To evaluate the reference background SNP for A549 cells, the authors loaded end-matched total RNA-seq data for A549 cells from ENCODE (28) (selected name: ENCFF542FVG, ENCFF538ZTA, ENCFF214JEZ, ENCFF629LOL, ENCFF149CJD, ENCFF006WNO, ENCFF828WTU, ENCFF380V). Each paired-end fastq file was first trimmed using the trim_galore/0.4.1 (51) adapter program with default parameters aligned to the hgl9 human genome construct using STAR/v2.5.2b {49). Unmapped reads and many mapped reads were removed using samtools/vl.3 (52). Duplicated reads were filtered out using the MarkDuplicates function in picard/1.105 (53). Non-duplicated reads from all samples were pooled and sorted using samtools/vl.3 (52). Background SNPs were requested using the mpileup function in samtools/vl.3 (52) and the mpileup2snp function in VarScan/2.3.9 {54). For the HEK2 93T and NIH/3T3 testing experiment, a background reference SNP was generated in a similar flow-type algorithm along with pooled single cell sam data under control conditions (under conditions without S4U labeling and without IAA treatment).

[00560] To create a single cell alignment file, all mutations with a quality score ≤13 were removed. Mutations at both ends of each read were due mainly to sequencing errors and were therefore also filtered out. For each read, the authors evaluated the presence of T>C mutations (for the sense strand) or A>G mutations (for the antisense strand), and these mutated reads were labeled as newly synthesized reads.

[00561] Each cell was characterized by two digital gene expression arrays based on full sequencing data and newly synthesized RNA data as described above. Genes with expression in 5 or less than 5 cells were filtered out. Cells with less than 2000 UMI or more than 80000 UMI were discarded. Cells with a doublet score >0.2 according to the Scrublet/0.2 (55) streaming type doublet analysis algorithm were removed.

[00562] Dimensional data were first reduced by PCA (after selecting the best 2000 genes with the highest variance) on digital gene expression arrays based on total gene expression data or newly synthesized gene expression data using Monocle 3 (55,57). The top 10 PCs were selected for dimensionality reduction analysis using the Uniform Multiple Approximation and Projection Program (UMAP/0.3.2), i.e., a newly proposed algorithm based on Riemannian geometry and algebraic topology for dimensionality reduction and data visualization (26) . For joint analysis, the authors pooled the best 10 PC calculated for the entire transcriptome and the best 10 PC calculated for the newly synthesized transcriptome for each individual cell before UMAP downsizing. Cell clusters were created using the densityPeak algorithm implemented in Monocle 3 (56, 57). First, the authors performed a UMAP analysis based on a total of all treated cells, and this cluster was identified as an outlier (724 out of 7404 cells). These cells were labeled with GATA3, a marker of highly expressed differentiated cells (34), and filtered before further analysis.

[00563] Transcription Factor (TF) Binding Assay to Regulatory Genes

[00564] The authors set out to identify relationships between TF and regulatory genes based on their covariance. Cells with more than 10,000 UMI were detected, and genes with newly synthesized reads detected in more than 10% of all cells were selected. Whole gene expression and number of newly synthesized genes per cell were normalized to cell-specific library size factors calculated with a whole gene expression matrix using Monocle 3's estimateSizeFactors algorithm (56, 57) and then log-transformed, centered, and scaled with using the scale() function in R. For each detected gene, a LASSO regression model was built using the glmnet software package (58) to predict normalized expression levels from the normalized expression of 853 TF annotated in the "motifAnnotations_hgnc" data from the RcisTarget software package {29) , by building the following model:

[00565] G _i =β ₀ +β _t T _i

[00566] where G _i is the corrected expression value of gene i. This value was calculated by counting the number of newly synthesized mRNAs for each cell, normalized by estimating cell-specific cell size factor (SG _i ) using estimateSizeFactors in Monocle 3 (56, 57) on a matrix for total expression in each cell, and subjected to a log transformation :

[00567]

[00568] To simplify the subsequent comparison of genes, the authors standardized the response G _i before building a model for each gene i with the scale () function in R.

[00569] Similarly to G _i , T _i is the adjusted value of TF expression for each cell. This value was calculated by calculating the total level of expression of TF for each cell, normalized by estimating the cell-specific cell size factor (SG _i ) using estimateSizeFactors in Monocle 3 [56, 57) on the matrix for total expression in each cell, and subjected to a logarithmic transformation :

[00570]

[00571] Before plotting, Ti is standardized using the scale() function in R.

[00572] The method of the invention targets TFs that can regulate each gene by looking for a subpopulation that can be used to predict its expression in a regression model. However, TF with expression correlated with gene expression does not guarantee that it will regulate that gene: if gene A is specifically expressed in a cell in state 1, and TF B is specifically expressed in cell type 2. Although negative correlations between TF expression and the rate of new gene synthesis may indicate transcriptional repressor activity, however, the authors felt that a more likely explanation for the negative associations reported by glmnet is the presence of mutually exclusive expression patterns specific to cell state and TF activity. Thus, at the time of prediction, the authors excluded TF, whose expression was negatively correlated with the level of gene synthesis, as well as with a low correlation coefficient (≤0.03). The authors identified a total of 6103 links between TF and regulatory genes.

[00573] In order to identify putative targets for direct binding, the authors chose the point of intersection of the bonds with TFs characterized in the ENCODE Chip-seq experiment (28). Of the 1086 TF associations characterized in ENCODE, 807 associations were confirmed at TF binding sites near gene promoters {59), which increased the odds ratio (number of confirmed associations versus unconfirmed associations) by a factor of 4.3 compared to background (probability ratio = 2.89 for associations identified by LASSO regression and 0.67 for background, p<2.2e-16, Fisher's exact test). Only sets of genes with significant enrichment of TF Chip-seq-only binding sites (Fischer's exact test, 5% false detection rate) were retained, and these sets were reduced to remove indirect target genes without retaining TF binding data. In this method, 591 relationships were saved.

[00574] To further confirm the associations of TF genes, the authors also used the SCENIC software package (29), i.e., a flow-type algorithm for constructing networks of regulatory genes based on the enrichment of TF target motifs around gene promoters (10 kb ). Each co-expression module identified by LASSO regression was analyzed by cis regulatory motif analysis using RcisTarget(29). Only modules with significant enrichment of the correct TF regulator motifs were retained and reduced to remove indirect target genes without retaining the motif. The authors filtered the associations of the TF gene by three correlation coefficient thresholds (0.3, 0.4, and 0.5) and pooled all associations confirmed with RcisTarget(29). A total of 509 connections were confirmed by the method of analysis of motives. After combining both methods, a total of 986 TF regulatory genes were identified by the covariance between TF expression level and gene synthesis as indicated by DNA binding data or motif analysis. To assess the likelihood that these associations would be artifacts of the adjusted regression, the authors permuted the matrix sample identifiers for TF expression and performed the same analysis. After such a rearrangement, no connections were found.

[00575] Ordering of cells by TF functional modules

[00576] To calculate TF activity in each cell, the numbers of newly synthesized UMIs for genes in the TF target module were scaled by library size, log-transformed and pooled, and then Z-score mapped. Since it has been suggested that TFs with highly correlated or uncorrelated activities can function in a biological process in a bound form, the authors calculated the absolute Pearson correlation coefficient between each pair of TF activities, and based on this clustered the TFs using the ward.d2 clustering method. in the software package pheatmap/1.0.12(60). Five TF functional modules have been identified and annotated based on their functions.

[00577] To characterize cell states by the size of each functional TF module, cells were ordered by activity of TF associated with the cell cycle (TF module 1) or TF associated with the GR response (TF module 3) using the UMAP method (metric = " cosine", n_neighbords = 30, min_dist = 0.01). The cell cycle trajectory was confirmed by cell cycle gene markers in Seurat/2.3.4(27). The three phases of the cell cycle were identified using the densityPeak algorithm implemented in Monocle 3 (56,57) by UMAP coordinates ordered by cell cycle TF modules. Because each major cell cycle phase still retained variable TF activity and cell cycle marker expression, we segmented each phase into early/middle/late states using k-means clustering (k=3) and obtained a total of nine cell cycle states. . Three states of the GR response have been identified using the densityPeak algorithm implemented in Monocle 3 (56,51).

[00578] Restoration of the previous state of the transcriptome by sci fate

[00579] To identify the former state of the transcriptome (the state of the cell before S4U labeling), we hypothesized that the mRNA half-life is constant under various DEX treatment conditions. This assumption was further confirmed by a subsequent stability test. According to this assumption, the total transcriptome partially disrupted 2 hours prior to S4U labeling should be the same in non-DEX-treated cells and in DEX-treated cells 2 hours later. Thus, the differences in the total transcriptome (by volume) should match the differences in the newly synthesized transcriptome (by volume), adjusted for the degree of detection depending on the technique:

[00580] A _0h /S _0h - (N _0h /S _0h )/α=A _2h /S _2h ~ (N _2h /S _2h )/α

[00581] A _0h means the number of pooled UMIs for all cells in the group without DEX treatment; S _0h means the size of the library (total number of cells for UMI) in the absence of DEX processing; N _0h means the number of pooled newly synthesized UMI for all cells in the group without DEX treatment; A _2h is the number of pooled UMIs for all cells in the DEX treatment group after 2 hours; S _2h is the library size (total UMI for cells) in the DEX treatment group after 2 hours; N2h is the number of pooled newly synthesized UMIs for all cells in the DEX treatment group after 2 hours; a means the degree of detection of sci-fate. Theoretically, one degree of detection can be determined for each gene. However, for genes with little difference in the rate of new synthesis between the two states, the estimated parameter a is within the noise level. Thus, the authors selected genes showing higher differences in the normalized rate of new synthesis between the two states: first, a series of gene filtering thresholds were tested and α calculated for each gene. Then the authors plotted the relationship between the threshold value and the ratio of genes with out-of-bounds values of α, (<0 or>1). The authors chose a threshold that, at the inflection point on the graph, corresponded to 186 selected genes. Differences in newly synthesized mRNA for these genes strongly correlate with differences in mRNA expression level (Pearson's test, r=0.93, Fig. 37A), suggesting that the level of detection of new RNA is fairly stable across all genes. Thus, the median rate of uptake of newly synthesized RNA by sci-fate is 82%.

[00582] The degree of mRNA degradation was then calculated for every 2 hours. Since the A549 cell population can be considered stable without external perturbation, for cells after 2 hours of DEX treatment, their previous state (2 hours before labeling with S4U) should be the same as for cells not treated with DEX. Similarly, the previous state (before labeling with S4U) for cells treated with DEX at time T=0/2/4/6/8/10 hours should be similar to the state of profiled cells at time T=0/0/2/4/6 /8 ocloc'k.

[00583] A _t1 /S _t1 - (N _t1 /S _t1 )/α=A _t0 /S _t0 ⋅β

[00584] A _t1 means the number of combined UMIs for all cells at t1; S _t1 means the size of the library (total number of cells for UMI) per t1; N _t1 is the number of newly synthesized UMIs pooled for all cells at t1; α means the estimated degree of detection of sci fate; A _t0 means the number of combined UMIs for all cells on t0; S _t0 means library size (total number of UMIs for cells) per t0; β=1 is the degree of gene-specific degradation for time t0-t1, and corresponds to the mRNA half-life γ, where β is calculated by the formula:

[00585] β=1-(1/2) ^(t1-t0)/γ

[00586] The degree of degradation of the β gene can be calculated after each DEX treatment at a 2-hour interval. After the aforementioned stability check, it was found that the degree of gene degradation was highly correlated at various DEX treatment times (FIG. 35B). Then, the authors used the average degree of gene degradation for further analysis.

[00587] Based on the estimated degree of detection and degradation of genes, the previous state of the transcriptome of each cell can be calculated by the formula:

[00588] a _t1 -n _t1 /α=a _t0 ⋅β

[00589] a _t1 means the number of UMI for individual cells at t1; n _t1 means the number of newly synthesized UMI for individual cells per t1; a means the estimated degree of detection of sci fate; β=1 - the degree of gene-specific decomposition for the time t0-t1. a _t0 means the estimated number of UMIs for individual cells over the past time point t0, where all negative values have been adjusted to 0.

[00590] Binding analysis to build a single cell state trajectory.

[00591] Through the analysis of binding, the authors have set the goal to identify associated parent and child cells in the same trajectory. Technically, for cells at t1, the authors combined the previous state of the transcriptome (before labeling with S4U, 2 hours before t1 in the experiment conducted by the authors) into one group 1, and the state of the entire transcriptome at t0 (2 hours before t1) into the second group 2. Assuming that there was no overt cell apoptosis, these two groups should have a similar distribution of cellular states. The authors applied a multiple alignment strategy to identify common cellular conditions between the two datasets based on common sources of variability (27). This analysis is based on the other assumption that the previous and current state of each cell (except cells at the start and end time points) are detected in detail, which is confirmed by the data sets obtained by the authors as more than 6000 cells are characterized (more than 1000 cells per condition), or one cells within less than one minute during the passage of the cell cycle. After the introduction of the flow-type algorithm, cell states starting from t0 and former states starting from t1 are aligned in the same UMAP space. The invalidity of the above assumptions can be detected by the outliers during the alignment of the two datasets. For each cell A at t1, the authors chose its nearest neighbors at t0 as the parent state when aligned in UMAP space. Similarly, for each cell at t0, the authors chose its nearest neighbors at t1 as the states of child cells. It should be noted that communication does not have to be bidirectional: the parent state of one cell can be associated with the state of another child cell. Since the parent state and child state were identified for each cell (except for cells at 0 o'clock and 10 o'clock), the authors then identified the associated parent cell of the parent of each cell and the similarly associated child cell of the descendant of each cell. Thus, each individual cell can be characterized by the path of the cell's transition from one state to another at all five time points covering a period of 10 hours. Since many cells (>50) were characterized in each state, a stochastic process of cell transition from one state to another can also be recorded.

[00592] Dimensionality reduction and clustering analysis to evaluate individual cell transcriptome dynamics.

[00593] To reduce dimensionality in individual cell transcriptome dynamics, for each state, the top 5 PCs for the entire transcriptome and the top 5 PCs for the newly synthesized transcriptome were selected, and these PCs were pooled in temporal order for each individual cell state trajectory for UMAP analysis. The main types of cell trajectories were identified using the density peak clustering algorithm {61).

[00594] Taking into account the ratio of the state of the cell at the initial moment of time (processing at 0 hours) and the probability of the transition of the cell from one state to another, estimated from the data obtained, the authors first predicted the distribution of cell states after 2 hours, if we assume that the process the transition of a cell from one state to another during DEX processing is a Markov dynamics, which is autonomous for cells and does not depend on time. Similarly, the distribution of cell states at a later point in time can be calculated based on the predicted distribution of cell states 2 hours before.

[00595] Predicting the probability of a cell transition from one state to another based on state instability

[00596] Cell state instability is defined as the probability of each state transitioning to other states after 2 hours. To calculate the distance of transitions between cells, the authors first selected an equal number (n=50) of cells in each state and combined the complete transcriptome and the newly synthesized transcriptome for all cells in this state. Each state of the cell can be determined by common information that combines the entire and newly synthesized transcriptome. The distance between cell states is calculated as the Pearson correlation coefficient for the total information between two states.

[00597] To predict the probability of transition between states, the authors built a 3-layer neural network (number of elements: 128, 128, 26 with resistance activation on each layer; loss function: cosine_proximity, batch size: 128, periods: 80) using Keras/2.2.4(62) programs. For input, the authors used the instability of the state at the current moment, the normalized instability of the state for the other 26 states (scaled by the instability of the current state), and the transition distance (square) from the current state to the other 26 states (in the same order of states as in the state instability vector ). To avoid over-approximation of the model, the authors permuted the order of instability states 200 times for each input, and the order of transition distances between states was maintained the same as in the case of instability states. To evaluate the performance of the model, the authors applied a hold/no hold function data validation program by training 26 states, and validating the model in the previous state, predicting the probabilities of state transitions to all other 26 states. To predict the transition probability from the transition distance only, the same training and validation model is used, with all initial state instabilities replaced by 1.

[00598] Bibliography

[00599] 1. N. Moris, C. Pina, A. M. Arias, Transition states and cell fate decisions in epigenetic landscapes. Nat. Rev. Genet. 17, 693-703 (2016).

[00600] 2. A. Filipczyk et al., Network plasticity of pluripotency transcription factors in embryonic stem cells. Nat. Cell biol. 17, 1235-1246 (2015).

[00601] 3. S. Hormoz et al., Inferring Cell-State Transition Dynamics from Lineage Trees and Endpoint Single-Cell Measurements. Cell System. 3, 419-433.e8 (2016).

[00602] 4. V. A. Herzog et al., Thiol-linked alkylation of RNA to assess expression dynamics. Nat. methods. 14, 1198-1204 (2017).

[00603] 5. J. A. Schofield, E. E. Duffy, L. Kiefer, M. C. Sullivan, M. D. Simon, TimeLapse-seq: adding a temporal dimension to RNA sequencing through nucleoside recoding. Nat. methods. 15, 221-225 (2018).

[00604] 6. J. C. Buckingham, Glucocorticoids: exemplars of multi-tasking. Br. J Pharmacol. 147, S258 (2006).

[00605] 7. M. D. Cleary, C. D. Meiering, E. Jan, R. Guymon, J. C. Boothroyd, Biosynthetic labeling of RNA with uracil phosphoribosyltransferase allows cell-specific microarray analysis of mRNA synthesis and decay. Nat. Biotechnol. 23, 232-237 (2005).

[00606] 8. L. Dolken et al., High-resolution gene expression profiling for simultaneous kinetic parameter analysis of RNA synthesis and decay. RNA. 14, 1959-1972 (2008).

[00607] 9. C. Miller et al., Dynamic transcriptome analysis measures rates of mRNA synthesis and decay in yeast. Mol. Syst. Biol. 7, 458-458 (2014).

[00608] 10. E. E. Duffy et al., Tracking Distinct RNA Populations Using Efficient and Reversible Covalent Chemistry. Mol. cell. 59, 858-866 (2015).

[00609] 11. B. Schwalb et al., TT-seq maps the human transient transcriptome. Science. 352, 1225-1228 (2016).

[00610] 12. M. Rabani et al., Metabolic labeling of RNA uncovers principles of RNA production and degradation dynamics in mammalian cells. Nat. Biotechnol. 29, 436-442 (2011).

[00611] 13. M. R. Miller, K. J. Robinson, M. D. Cleary, C. Q. Doe, TU-tagging: cell type-specific RNA isolation from intact complex tissues. Nat. methods. 6, 439-441 (2009).

[00612] 14. D. A. Cusanovich et al., Multiplex single cell profiling of chromatin accessibility by combinatorial cellular indexing. Science. 348, 910-914 (2015).

[00613] 15. J. Cao et al., Comprehensive single-cell transcriptional profiling of a multicellular organism. Science. 357, 661-667 (2017).

[00614] 16. J. Cao et al., Joint profiling of chromatin accessibility and gene expression in thousands of single cells. Science. 361, 1380-1385 (2018).

[00615] 17. V. Ramani et al., Massively multiplex single-cell Hi-C (2016), doi:10.1101/065052.

[00616] 18. R. M. Mulqueen et al., Highly scalable generation of DNA methylation profiles in single cells. Nat. Blotechnol. 36, 428-431 (2018).

[00617] 19. S. A. Vitak et al., Sequencing thousands of single-cell genomes with combinatorial indexing. Nat. methods. 14, 302-308 (2017).

[00618] 20. Y. Yin et al., High-throughput mapping of meiotic crossover and chromosome mis-segregation events in interspecific hybrid mice (2018), doi:10.1101/338053.

[00619] 21. A. B. Rosenberg et al., Single-cell profiling of the developing mouse brain and spinal cord with split-pool barcoding. Science. 360, 176-182 (2018).

[00620] 22. T. E. Reddy et al., Genomic determination of the glucocorticoid response reveals unexpected mechanisms of gene regulation. Genome Res. 19, 2163-2171 (2009).

[00621] 23. S. John et al., Chromatin accessibility predetermines glucocorticoid receptor binding patterns. Nat. Genet. 43, 264-268 (2011).

[00622] 24. T. E. Reddy, J. Gertz, G. E. Crawford, M. J. Garabedian, R. M. Myers, The Hypersensitive Glucocorticoid Response Specifically Regulates Period 1 and Expression of Circadian Genes. Mol. cell. Biol. 32, 3756-3767 (2012).

[00623] 25. C.M. Vockley et al., Direct GR Binding Sites Potentiate Clusters of TF Binding across the Human Genome. cell. 166, 1269-1281.el9 (2016).

[00624] 26. L. McInnes, J. Healy, N. Saul, L. GroBberger, UMAP: Uniform Manifold Approximation and Projection. Journal of Open Source Software. 3, 861 (2018).

[00625] 27. A. Butler, P. Hoffman, P. Smibert, E. Papalexi, R. Satija, Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat. Biotechnol. 36, 411-420 (2018).

[00626] 28. The ENCODE Project Consortium, The ENCODE (ENCyclopedia Of DNA Elements) Project. Science. 306, 636-640 (2004).

[00627] 29. S. Aibar et al., SCENIC: single-cell regulatory network inference and clustering. Nat. methods. 14, 1083-1086 (2017).

[00628] 30. M. Boruk, J. G. A. Savory, R. J. G. Hache, AF-2-Dependent Potentiation of CCAAT Enhancer Binding Proteinβ - Mediated Transcriptional Activation by Glucocorticoid Receptor. Mol. Endocrinol. 12, 1749-1763 (1998).

[00629] 31. W. Qin et al., Identification of functional glucocorticoid response elements in the mouse FoxO1 promoter. Biochem. Biophys. Res. commun. 450, 979-983 (2014).

[00630] 32. C. S. Sheela Rani, N. Elango, S.-S. Wang, K. Kobayashi, R. Strong, Identification of an Activator Protein-1-Like Sequence as the Glucocorticoid Response Element in the Rat Tyrosine Hydroxylase Gene. Mol. Pharmacol. 75, 589 (2009).

[00631] 33. M. Fischer, G. A. Muller, Cell cycle transcription control: DREAM/MuvB and RB-E2F complexes. Crit. Rev. Biochem. Mol. Biol. 52, 638-662 (2017).

[00632] 34 J. Chou, S. Provot, Z. Werb, GATA3 in development and cancer differentiation: GATA cells have it! J. Cell. physiol. 222, 42-49 (2010).

[00633] 35. J. Y. C. Madhurima Biswas, Role of Nrfl in antioxidant response element-mediated gene expression and beyond. Toxicol. Appl. Pharmacol. 244, 16 (2010).

[00634] 36. I.-G. Ryoo, M.-K. Kwak, Regulatory crosstalk between the oxidative stress-related transcription factor Nfe212/Nrf2 and mitochondria. Toxicol. Appl. Pharmacol. 359, 24-33 (2018).

[00635] 37. J. M. Harmon, M. R. Norman, B. J. Fowlkes, E. B. Thompson, Dexamethasone induces irreversible G1 arrest and death of a human lymphoid cell line. J. Cell. physiol. 98, 2 67-278 (1979).

[00636] 38. A. K. Greenberg et al., Glucocorticoids inhibit lung cancer cell growth through both the extracellular signal-related kinase pathway and cell cycle regulators. Am. J. Respir. Cell Mol. Biol. 27, 320-328 (2002).

[00637] 39. J. Cao et al., Comprehensive single-cell transcriptional profiling of a multicellular organism. Science. 357, 661-667 (2017).

[00638] 40. J. Cao et al., Joint profiling of chromatin accessibility and gene expression in thousands of single cells. Science. 361, 1380-1385 (2018).

[00639] 41. W. Matsushima et al., SLAM-ITseq: sequencing cell type-specific transcriptomes without cell sorting. Development. 145 (2018), doi:10.1242/dev.164640.

[00640] 42. U. Sharma et al., Small RNAs are trafficked from the epididymis to developing mammalian sperm (2017), doi:10.1101/194522.

[00641] 43. A. McKenna et al., Whole-organism lineage tracing by combinatorial and cumulative genome editing. Science. 353, aaf7907 (2016).

[00642] 44. B. Raj et al., Simultaneous single-cell profiling of lineages and cell types in the vertebrate brain. Nat. Biotechnol. 36, 442-450 (2018).

[00643] 45. K. L. Frieda et al., Synthetic recording and in situ readout of lineage information in single cells. Nature. 541, 107-111 (2017).

[00 644] 46. H. Wickham, ggplot2: Elegant Graphics for Data Analysis (Springer, 2016).

[00645] 47. M. Muhar et al., SLAM-seq defines direct gene-regulatory functions of the BRD4-MYC axis. Science. 360, 800-805 (2018).

[00646] 48. J. Cao et al., Comprehensive single-cell transcriptional profiling of a multicellular organism. Science. 357, 661-667 (2017).

[00647] 49. A. Dobin et al., STAR: ultrafast universal RNA-seq aligner. bioinformatics. 29, 15-21 (2013).

[00648] 50. P. Lindenbaum, JVarkit: java-based utilities for Bioinformatics. figshare (2015).

[00649] 51. FelixKrueger, FelixKrueger/TrimGalore. GitHub, (available at https://github.com/FelixKrueger/TrimGalore).

[00650] 52. H. Li et al., The Sequence Alignment/Map format and SAMtools. bioinformatics. 25, 2078-2079 (2009).

[00651] 53. Picard Tools - By Broad Institute, (available at http://broadinstitute.github.io/picard/).

[00652] 54. D. C. Koboldt et al., VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Res. 22, 568-576 (2012).

[00653] 55. S. L. Wolock, R. Lopez, A. M. Klein, Scrublet: computational identification of cell doublets in single-cell transcriptomic data (2018), doi:10.1101/357368.

[00654] 56. X. Qiu et al., Reversed graph embedding resolves complex single-cell trajectories. Nat. methods. 14, 979-982 (2017).

[00655] 57. cole-trapnell-lab, cole-trapnell-lab/monocle-release. GitHub, (available at https://github.com/cole-trapnell-lab/monocle-release).\

[00656] 58. J. Friedman, T. Hastie, R. Tibshirani, Regularization Paths for Generalized Linear Models via Coordinate Descent. J. Stat. softw. 33 (2010),

doi:10.18637/jss.v033.i01.

[00657] 59. Dataset - ENCODE Transcription Factor Binding Site Profiles, (available at http://amp.pharm.mssm.edu/Harmonizome/dataset/ENCODE+Transcription+Factor+Binding+Site+Profiles).

[00658] 60. raivokolde, raivokolde/heatmap.GitHub, (available at https://github.com/raivokolde/heatmap).

[00659] 61. A. Rodriguez, A. Laio, Clustering by fast search and find of density peaks. Science. 344, 1492-1496 (2014).

[00660] 62. keras-team, keras-team/keras. GitHub, (available at https://github.com/keras-team/keras

[00661] Example 4

[00662] Multiplex Transcript Capture

[00663] Most single cell RNA sequencing methods are performed at a coverage of 15,000 to 50,000 unique reads per cell (Ziegenhain et al. 2017), while the total mRNA content in single cells can vary from 50,000 to 300,000 molecules (Marinov et al. 2014) . In addition, most of these methods use reverse transcription (RT) oligo(dT) primers and these methods aim at sequencing at the 3' end of the RNA. This means that such methods have limited ability to detect changes in the abundance of any given transcript. Recent studies characterizing large numbers of cells (Gasperini et al. 2019; Cao et al. 2019) have required very high sequencing depths, with the Illumina NovaSeq programs used in these studies costing $30,000 each to run, making such experiments inaccessible to most scientists.

[00664] However, in both cases, the number of read operations required to obtain biological information from the available data is relatively small. When obtaining information about noncoding perturbations in individual cells, analysis for changes in the expression level was performed only for genes located in the cis position relative to the disrupted regulatory element (Xie et al. 2017; Gasperini et al. 2018). Although common expression patterns were used in cell atlas experiments to cluster similar cells, assignment to specific cell types was performed using a small number of key transcription factor genes. Thus, the ability to obtain readout data for gene transcripts that are most informative in these experiments can significantly reduce the required sequencing depth and increase the power of analysis to detect minor differences between cells.

[00665] The authors focused on single cell sequencing to identify mRNAs of interest using specific RT primers rather than oligo(dT) primers. A similar method has recently been used to basically sequence-specific all known splice junctions in yeast, resulting in a 100-fold enrichment in target regions compared to non-target regions (Xu et al., 2018). A pool of RT primers that span the transcripts of interest reduces the number of transcriptome library (sci-PHK-seq) reads to hundreds of captured transcripts per experiment.

[00666] These sciPHK-seq manipulations have many advantages over oligo-(dT)-priming. First, these manipulations allow for direct sequencing of genome regions that were identified by the authors as the most informative for each experiment. Secondly, they allow reverse transcription of each RNA molecule into cDNA, which increases the probability of detection per RNA molecule. Third, this approach allows targeting only those amplicons that have been uniquely mapped and can reduce the background of ribosomal reads to a greater extent than alternative methods using randomized hexamer or oligo(dT)-priming. Fourth, it allows targeting of informative regions of mRNA, such as splicing junctions and exons resulting from alternative events at transcription initiation sites, and thus obtain isoform information that is not easily detected using conventional sciRNA-seq.

[00667] sciPHK-seq is ideal for modification with multiple OT primers. Most single-cell RNA-seq methods use spheres associated with unique oligo-identifiers to introduce cell-identifying barcodes into each cell's transcriptome, typically by capturing mRNA via hybridization to their poly(A) tail. Although such spheres have been modified to add a small number of specific RT primers in order to increase the coverage of multiple transcripts (Saikia et al. 2018), however, this strategy will be difficult to scale up to hundreds of target transcripts or make quick swaps between experiments. Thus, the adaptability of combinatorial indexing of single cells will be useful in the development of multiplex RT-sequencing of RNA-seq in single cells.

[00668] The workflow for this aspect is similar to the three-layer sciPHK-seq protocol described in Examples 1 and 3, but in some embodiments, it does not include an OT step.

[00669] 1. Construction of a pool of OT primers. In one aspect, these primers will be synthesized separately and then combined. To target >384 amplicons, a library of primers amplified as double-stranded DNA can be synthesized, followed by processing to obtain single-stranded primers as described in the literature (Xu et al. 2018). This second strategy allows multiple unique indexes to be added to the RT primers (allowing sciPHK-seq to be indexed in RT and final PCR).

[00670] 2. Conducting multiplex OT using primer pool. This reaction can either be a single reaction with thousands of cells (if indexing was not performed at that stage) or multiple parallel reactions that will add a highly specific index on reverse transcription.

[00671] 3. Ligation of the hairpin adapter to add a well-specific index.

[00672] 4. Combining all cells and performing second strand synthesis.

[00673] 5. Allocation of cells to a plurality of wells and labeling to add a second persistent PCR fragment.

[00674] 6. PCR amplification, addition of a well-specific final index.

[00675] 7. Sequencing.

[00676] Primer construction workflow:

[00677] 1. Collection of sequences for all exons from target genes.

[00678] 2. Computer analysis of all possible 25 bp RT primers.

[00679] 3. Filtering candidate RT primers by:

[00680] a. a GC content of 40-60% corresponding to melting points of approximately 55 to 70 degrees.

[00681] b. at least 2 Gs or Cs in the last 5 nucleotides of the primer to increase the likelihood that the hybridized RT primer will be a good substrate for reverse transcriptase enzyme extension.

[00682] p. probabilities of non-target priming. In the first experiment according to the invention, the authors found that although the target genes according to the invention were highly enriched, however, most of the reads still came from other RNAs that are present in cells in excess. Most of these non-targeted priming events were the result of ~5-8 bp. - complementarity between the 3' end of the primer and the non-target RNA. Thus, the primer design algorithm developed by the authors was made taking into account the excess of the final hexamer of the RT primer in total cellular RNA. The authors included only those OT primers in which this last hexamer:

[00683] i. Generally not present in ribosomal RNA. This hexamer is derived from a series of previously described hexamers that “have not been appropriately randomized, or NSR hexamers (Armour et al. 2009). Primers that pass through this filter are much less likely to be non-target primers in ribosomal RNA.

[00684] ii. Present in total cellular RNA in low amounts. The authors counted all 4096 possible hexamers in PRO-seq reads mapped by the human genome (Core et al., 2014). PRO-seq measures increasing transcription in cells, including ribosomal transcription. The authors used only OT primers that end in hexamers present in the lower quartile of excess in this data set. This makes it possible to preserve several hexamers, which, although present in ribosomal RNA, are not as common in cells as RNA.

[00685] This excess content filter drastically changes primer selection. There is only ~17% overlap between primers selected using the streaming type algorithm of the invention in the presence or absence of this filter. In the future, variants of the flow type construction algorithm developed by the authors will improve the filtering of non-target objects. As data are collected for more primers, authors should be able to evaluate more non-target priming events.

[00686] 4. Candidate filtering by mapability. The authors aligned each candidate for hgl9 using a bow tie program, which allowed 3 mismatches to be identified. This step ensures that each primer has only one target site in the genome.

[00687] 5. Selecting from the candidate primers that have passed through these filters, the set that is most evenly distributed throughout the gene.

[00688] For each gene that is targeted, the authors decide how many primers to design per exon. The authors included the first and last primers that pass the filters for each exon, and then assembled the inner primers that most evenly cover the exon by minimizing the distance from the primer positions that would accurately divide the exon into n chunks.

[00689] So, for example, for the 300 bp exon, where the authors searched for 3 primers, the primers closest to positions 1, 150, and 300 were taken, which passed all filters up to this point.

[00690] 6. For this pilot experiment, the RT primers were ordered in 384-well plates and pooled to create an equimolar mixture of all primers. This mixture was then phosphorylated with T4 polynucleotide kinase to ligate the indexed hairpin oligonucleotide during the creation of the sciRNA-seq library (Cao et al., 2019). This is much more advantageous than ordering phosphorylated oligonucleotides. OT primers in 2 5 p. also add a unique molecular identifier (UMI) in 8 bp. and fragments in 6 p. to anneal a hairpin oligonucleotide that adds a well-specific index (for combinatorial indexing) and a PCR fragment.

[00691] This process can be iterative, with each RT primer ordered separately: a lower off-target ratio was achieved in later experiments by selectively recombining primers found to have favorable uptake rates in the first experiment. Each Illumina sequencing read covers a 25 bp RT primer. and a captured RNA molecule, allowing RT primers and captured molecules to be mapped separately to calculate the targeting rate for each primer.

[00692] Later rounds can include more OT primers by synthesizing them in an array. The primer library can be propagated by PCR and made single stranded by selective exonucleolytic strand degradation that does not include a blocking group in the PCR primer (Xu et al., 2018). A large array can be used to synthesize several pools of primers: if each pool has a specific PCR fragment, then one array can be used to generate tens of pools of thousands of primers, each of which can be selectively amplified.

[00693] Multiplex reverse transcription:

[00694] Multiplex target capture can be carried out in several stages in accordance with the protocol for creating a PHK-seq library. However, the authors believe that reverse transcription is most easily carried out in parallel. Highly multiplex PCR reactions are rarely successful. PCR reactions include many (10-20) cycles. This means that problems with off-target annealing are exacerbated after exponential growth during these cycles, which often outpace targeted annealing cycles. In multiplex PCR, two specific PCR primers are prepared for each target. The goal is for these two primers to specifically amplify only their target. However, in a large primer pool there will be several combinations that hybridize with other primers present in the pool. Since the concentration of primers is much higher than the concentration of template molecules, these primer dimers will dominate the pool by the end of the PCR. The inability to perform highly multiplexed PCR is the reason why many targeted amplification protocols, such as exome sequencing, often use molecular inversion probes to capture targets (Hiatt et al. 2013). In such protocols, target specificity is achieved with a single annealing step between probe and target. Target specific probes add PCR fragments, which are then used in the overall target PCR amplification. Single-cell combinatorial indexing methods rely on indexing in several steps during library creation: the inversion probe method for capturing targets from cDNA does not allow for a sufficient number of indexing steps.

[00695] To capture the multiplex target, the authors used a specific reverse transcription primer, after which they performed a PCR reaction that amplifies all molecules subjected to reverse transcription. Thus, the authors' strategy is similar to using molecular inversion probes for targeted DNA amplification: one step (reverse transcription) selectively targets transcripts of interest and adds a common PCR fragment that can be used to amplify all target molecules during PCR. Thus, high specificity in reverse transcription plays a very important role. Maintaining a high temperature after annealing the RT primers facilitates multiplex specific priming. Common reverse transcription protocols include denaturing the mixture of RNA and reverse transcription primer and cooling to 4 degrees to allow for annealing. This low annealing temperature is too mild for non-target annealing events. Authors should ensure that the only annealing events that can be extended are events for which all of the highly specific RT primers developed by the authors have found their targets. Thus, the authors maintained a high temperature throughout the entire protocol, as in other multiplex specific reverse transcription methods (Xu et al. 2018). The authors denatured a mixture of fixed cells, a pool of RT primers and dNTP at 65°C, annealed at 53°C, and then added a mixture of enzyme/reverse transcription buffer, which was pre-equilibrated at 53°C, to carry out the annealing reaction, and then an extension cycle was carried out at 53° C. for 20 minutes. Thus, the RT primers were unable to anneal at low temperature between the denaturation and extension steps.

[00696] The remaining steps of this method correspond to the methods described in Examples 1 and 3. The hairpin adapter is ligated in situ with the addition of a cell index. Cells are pooled, washed and distributed to new wells for the final indexing step. These wells carry out the synthesis of the second chain. Then, the double-stranded cDNA is labeled with the addition of a second common PCR fragment (the first fragment is obtained by ligation and the second by labeling). DNA is isolated from cells by binding to Ampure spheres followed by PCR with the addition of a second index.

[00697] Preliminary results:

[00698] All results shown in FIG. 40-42 are derived from a total library (no single cell combinatorial indexing) in situ (all steps performed in paraformaldehyde-fixed nuclei) generated using a pool of RT primers targeting genes at the LMO2 locus in K562 cells.

[00699] Bibliography:

[00700] Armour, Christopher D., John C. Castle, Ronghua Chen, Tomas Babak, Patrick Loerch, Stuart Jackson, Jyoti K. Shah, et al. 2009. "Digital Transcriptome Profiling Using Selective Hexamer Priming for cDNA Synthesis." Nature Methods 6(9): 647-49.

[00701] Cao, Junyue, Malte Spielmann, Xiaojie Qiu, Xingfan Huang, Daniel M. Ibrahim, Andrew J. Hill, Fan Zhang, et al. 2019. "The Single-Cell Transcriptional Landscape of Mammalian Organogenesis." Nature 566 (7745): 496-502.

[00702] Core, Leighton J., Andre L. Martins, Charles G. Danko, Colin T. Waters, Adam Siepel, and John T. Lis. 2014. "Analysis of Nascent RNA Identifies a Unified Architecture of Initiation Regions at Mammalian Promoters and Enhancers." Nature Genetics 46(12): 1311-20.

[00703] Gasperini, Molly, Andrew J. Hill, Jose L. McFaline-Figueroa, Beth Martin, Seungsoo Kim, Melissa D. Zhang, Dana Jackson, et al. 2019. "A Genome-Wide Framework for Mapping Gene Regulation via Cellular Genetic Screens." Cell 176(6): 1516.

[00704] Gasperini, Molly, Andrew Hill, Jose L. McFaline-Figueroa, Beth Martin, Cole Trapnell, Nadav Ahituv, and Jay Shendure. 2018. "crisprQTL Mapping as a Genome-Wide Association Framework for Cellular Genetic Screens." bioRxiv. https://doi.org/10.1101/314344.

[00705] Hiatt, Joseph B., Colin C. Pritchard, Stephen J. Salipante, Brian J. O'Roak, and Jay Shendure. 2013. "Single Molecule Molecular Inversion Probes for Targeted, High-Accuracy Detection of Low-Frequency Variation." Genome Research 23(5): 843-54.

[00706] Marinov, Georgi K., Brian A. Williams, Ken McCue, Gary P. Schroth, Jason Gertz, Richard M. Myers, and Barbara J. Wold. 2014. "From Single-Cell to Cell-Pool Transcriptomes: Stochasticity in Gene Expression and RNA Splicing." Genome Research 24(3): 496-510.

[00707] Saikia, Mridusmita, Philip Burnham, Sara H. Keshavjee, Michael F. Z. Wang, Pablo Moral-Lopez, Meleana M. Hinchman, Charles G. Danko, John S. L. Parker, and Iwijn De Vlaminck. 2018. "Simultaneous Multiplexed Amplicon Sequencing and Transcriptome Profiling in Single Cells." bioRxiv. https://doi.org/10.1101/328328.

[00708] Xie, Shiqi, Jialei Duan, Boxun Li, Pei Zhou, and Gary C. Hon. 2017. "Multiplexed Engineering and Analysis of Combinatorial Enhancer Activity in Single Cells." Molecular Cell 66(2): 285-99.e5.

[00709] Xu, Hansen, Benjamin J. Fair, Zach Dwyer, Michael Gildea, and Jeffrey A. Pleiss. 2018. "Multiplexed Primer Extension Sequencing Enables High Precision Detection of Rare Splice Isoforms." bioRxiv. https://doi.org/10.1101/331629.

[00710] Ziegenhain, Christoph, Beate Vieth, Swati Parekh, Bjorn Reinius, Amy Guillaumet-Adkins, Martha Smets, Heinrich Leonhardt, Holger Heyn, Ines Hellmann, and Wolfgang Enard. 2017. "Comparative Analysis of Single-Cell RNA Sequencing Methods." Molecular Cell 65(4): 631-43.e4.

[00711] Full disclosure of all patents, patent applications and publications cited herein, as well as materials available in electronic form (including, for example, representation of nucleotide sequences, for example, in GenBank and RefSeq, and representation of amino acid sequences, for example, in SwissProt, PIR , PRF, PDB, and translations from the annotated coding regions in GenBank and RefSeq) are incorporated herein by reference in their entirety. Additional materials cited in publications (such as additional tables, additional drawings, additional materials and methods, and/or additional experimental data) are also incorporated herein by reference in their entirety. In the event that there is any inconsistency between the disclosure of this application and the disclosure(s) of any document incorporated in this application by reference, the description of this application should take precedence. The foregoing detailed description and examples are provided only for a better understanding of the invention. However, they should not be construed as limiting the scope of the invention. The disclosure of the present invention is not limited to the specific details presented and described here, since it is obvious to a person skilled in the art that changes defined by the claims may be included in the present invention.

[00712] Unless otherwise stated, all numbers expressing amounts of components, molecular weights, and the like used in the specification and claims are to be understood as "approximate" in all cases. Accordingly, unless otherwise stated, the numerical parameters presented in the description and in the claims are approximate values that may vary depending on the desired properties to be achieved in the present invention. Each numerical parameter should be considered at least as a number denoted by significant figures using the usual methods of rounding, but, nevertheless, it should not be interpreted as an attempt to limit the doctrine of equivalents to the scope of the claims.

[00713] Although the numerical ranges and parameters defining the broad scope of the disclosure of the invention are approximate, however, the numerical values indicated in specific examples are reported with the greatest possible accuracy. That being said, all numerical values inherently contain an interval that is necessarily the result of the standard deviation found in their respective experimental measurements.

[00714] All headings are presented for the convenience of the reader and should not be construed as limiting the meaning of the text following the heading unless otherwise noted.

--->

SEQUENCE LIST

<110> ILLUMINA, INC.

UNIVERSITY OF WASHINGTON

<120> LARGE-SCALE MONOCELLULAR TRANSCRIPTOME LIBRARIES

AND METHODS OF THEIR OBTAINING AND APPLICATION

<130> IP-1722-PCT

<140> PCT/US2019/035422

<141> 2019-06-04

<150> 62/821.678

<151> 2019-03-21

<150> 62/680.259

<151> 2018-06-04

<160> 11

<170>PatentIn version 3.5

<210> 1

<211> 54

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<220>

<221> modified_base

<222> (7)..(24)

<223> a, c, t, g, unknown or other

<400> 1

cagagcnnnn nnnnnnnnnn nnnntttttt tttttttttt tttttttttt tttt 54

<210> 2

<211> 45

<212> DNA

<213> Artificial sequence

<220>

<223> Synthetic sequence description: synthetic

oligonucleotide

<220>

<223> Description of Combined DNA/RNA Molecule: Synthetic

oligonucleotide

<220>

<221> modified_base

<222> (7)..(16)

<223> a, c, t, g, unknown or other

<220>

<221> misc_feature

<222> (7)..(16)

<223> This area can include 9-10 bases

<220>

<221> modified_base

<222> (36)..(45)

<223> a, c, t, g, unknown or other

<220>

<221> misc_feature

<222> (36)..(45)

<223> This area can include 9-10 bases

<400> 2

gctctgnnnn nnnnnnuacg acgctcttcc gatctnnnnn nnnnn 45

<210> 3

<211> 29

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<400> 3

aatgatacgg cgaccaccga gatctacac 29

<210> 4

<211> 24

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<400> 4

caagcagaag acggcatacg agat 24

<210> 5

<211> 68

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<220>

<221> modified_base

<222> (19)..(36)

<223> a, c, t, g, unknown or other

<220>

<221> modified_base

<222> (68)..(68)

<223> a, c, t, g, unknown or other

<400> 5

acgacgctct tccgatctnn nnnnnnnnnn nnnnnntttt tttttttttt tttttttttt 60

68

<210> 6

<211> 29

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<400> 6

aatgatacgg cgaccaccga gatctacac 29

<210> 7

<211> 24

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<400> 7

caagcagaag acggcatacg agat 24

<210> 8

<211> 33

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<400> 8

acactctttc cctacacgac gctcttccga tct 33

<210> 9

<211> 15

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<400> 9

gtctcgtggg ctcgg 15

<210> 10

<211> 33

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<400> 10

acactctttc cctacacgac gctcttccga tct 33

<210> 11

<211> 15

<212> DNA

<213> Artificial sequence

<220>

<223> Artificial sequence description: synthetic

primer

<400> 11

gtctcgtggg ctcgg 15

<---

Claims (150)

1. A method for obtaining a library for sequencing containing nucleic acids originating from a plurality of individual nuclei or cells, where this method includes: (a) obtaining a plurality of nuclei or cells in a first plurality of compartments, where each compartment contains a subpopulation of nuclei or cells; (b) labeling newly synthesized RNA in subpopulations of cells or nuclei derived from cells; (c) processing the RNA molecules in each subset of nuclei or cells to obtain indexed nuclei or cells, where the processing includes adding to the RNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed DNA nucleic acids present in the indexed nuclei or cells, where the processing includes ligation, primer extension, hybridization or amplification; and (d) pooling indexed nuclei or cells to form pooled indexed nuclei or cells. 2. The method according to claim 1, where the processing includes: contacting subpopulations with a reverse transcriptase and a primer that hybridize to RNA nucleic acids to form primer-containing double-stranded DNA nucleic acids and the corresponding DNA nucleotide sequence from RNA template molecules, and optionally a) where the primer contains a poly-T nucleotide sequence that hybridizes to the poly(A)-tail of the mRNA, where optional processing also includes contacting subpopulations with a second primer, where the second primer contains a sequence that hybridizes to a predetermined DNA nucleic acid, where optionally, the second primer contains a compartment-specific index; b) where the primer contains a sequence that hybridizes to a predefined RNA nucleic acid, where optionally the method includes primers in different compartments that hybridize to different nucleotides of the same predefined RNA nucleic acid, or c) where the primer includes a template switch primer. 3. The method of claim 1, wherein the processing for adding the first compartment-specific index sequence comprises a two-step method of adding a nucleotide sequence containing the universal sequence to RNA nucleic acids to produce DNA nucleic acids, and then adding the first compartment-specific index sequence to nucleic acids. DNA acids. 4. A method of obtaining a library for sequencing containing nucleic acids originating from a plurality of individual nuclei or cells, where this method includes: (a) obtaining a plurality of nuclei or cells in a first plurality of compartments, where each compartment contains a subpopulation of nuclei or cells; (b) labeling newly synthesized RNA in subpopulations of cells or nuclei derived from cells; (c) contacting each subpopulation with a reverse transcriptase and a primer that hybridizes to a predetermined RNA nucleic acid to form primer-containing double-stranded DNA nucleic acids and the corresponding DNA nucleotide sequence from the RNA template molecules; (d) processing DNA molecules in each subset of nuclei or cells to obtain indexed nuclei or cells, where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed nucleic acids present in the indexed nuclei or cells, where the processing includes ligation, primer extension, hybridization or amplification; and (e) combining indexed nuclei or cells to form combined indexed nuclei or cells. 5. The method according to claim 4, where a) the primer contains the first compartment-specific index sequence, or b) where the processing to add the first compartment-specific index sequence includes a two-step method of adding a nucleotide sequence containing a universal sequence to the nucleic acids, and then adding the first compartment-specific index sequence to the nucleic acids. 6. The method according to claim 2 or 5, wherein the predetermined RNA nucleic acid is mRNA. 7. The method according to claim 1 or 5, where a) already existing RNA nucleic acids and newly synthesized RNA nucleic acids are labeled with the same indices in the same compartment, b) where the labeling includes the incubation of a plurality of nuclei or cells in a composition containing a nucleotide tag, where the nucleotide tag is included in the newly synthesized RNA, optionally i) where the nucleotide tag contains a nucleotide analog, a hapten-labeled nucleotide, a mutagenic nucleotide, or a nucleotide that can be modified by a chemical reaction, or ii) where more than one nucleotide tag is included in the newly synthesized RNA, where optionally the ratios of the nucleotide tag or tags are different for different compartments or time points, c) also involves treating subpopulations of nuclei or cells under predetermined conditions prior to labeling, optionally - where the predefined condition includes processing by the agent, optionally i) where the agent includes a protein, a non-ribosomal protein, a polyketide, an organic molecule, an inorganic molecule, an RNA or RNAi molecule, a carbohydrate, a glycoprotein, a nucleic acid, or a combination thereof, or ii) where the agent includes a therapeutic drug, - where the predetermined conditions of two or more compartments are different, - where processing and labeling is carried out simultaneously or processing is carried out before labeling. 8. The method according to claim 1 or 4, further comprising: distributing subpopulations of the combined indexed nuclei or cells over a second set of compartments and adding to the indexed nucleic acids present in the subpopulations of nuclei or cells a second index sequence to obtain nuclei or cells with two indexes containing nucleic acid fragments with two indexes, where the addition includes ligation, primer extension, hybridization, amplification or transposition; combining nuclei or cells with two indices to obtain combined nuclei or cells with two indices. 9. The method of claim 8, further comprising: distributing the subpopulations of fused nuclei or cells with two indexes over a third set of compartments and adding to the indexed nucleic acids present in the subpopulations of nuclei or cells with a third index sequence to obtain nuclei or cells with three indexes containing nucleic acid fragments with three indexes, where the addition includes ligation, hybridization, primer extension, amplification or transposition; combining nuclei or cells with three indices to obtain combined nuclei or cells with three indices, optional a) where the distribution includes breeding, b) where distribution includes sorting, c) where the addition comprises contacting the subpopulations with the hairpin ligation duplex under conditions suitable for ligating the hairpin ligation duplex to the end of nucleic acid fragments containing one or two index sequences, d) where the addition includes contacting nucleic acid fragments containing one or more index sequences with a transposome complex, where the transposome complex in the compartments contains a transposase and a universal sequence, where the contact also includes conditions suitable for fragmenting the nucleic acid fragments and incorporating the universal sequence into the fragments nucleic acid. 10. The method according to claim 1 or 4, where the addition includes ligation of the first compartment-specific index sequence, and also includes the addition of a second index sequence to obtain nuclei or cells with two indexes containing nucleic acid fragments with two indexes, where the addition includes transposition. 11. The method of claim 8, wherein the addition comprises ligation of a second compartment-specific index sequence, and also includes the addition of a third index sequence to produce two-index nuclei or cells containing three-index nucleic acid fragments, where the addition includes transposition. 12. The method according to any one of paragraphs. 1, 8 or 9, a) where the compartment includes a well or drop, b) where the compartments of the first set of compartments contain from 50 to 100,000,000 nuclei or cells, optionally according to claim 8, i) where the compartments of the second plurality of compartments contain from 50 to 100,000,000 nuclei or cells, ii) also includes obtaining nucleic acids with two indexes from the combined nuclei or cells with two indexes and thereby obtaining a library for sequencing from a plurality of nuclei or cells, optionally according to claim 9, i) where the compartments of the third set of compartments contain from 50 to 100,000,000 nuclei or cells, ii) also includes obtaining nucleic acids with three indexes from the pooled nuclei or cells with three indexes and thereby obtaining a library for sequencing from a plurality of nuclei or cells. 13. The method of claim 1 or 4, further comprising obtaining indexed nucleic acids from the pooled indexed nuclei or cells and thereby obtaining a sequencing library from the plurality of nuclei or cells. 14. The method according to any one of paragraphs. 1-11, further comprising: obtaining a surface containing a plurality of amplification sites, where the sites of amplification include at least two populations of linked single-stranded capture oligonucleotides having a free 3' end, and contacting a surface containing amplification sites with nucleic acid fragments containing one, two or three index sequences under conditions suitable to obtain a plurality of amplification sites, each containing a clonal population of amplicons, from a single fragment containing a plurality of indices. 15. The method according to any one of paragraphs. 8 or 9, wherein the addition of the compartment-specific index sequence comprises a two-step method of adding a nucleotide sequence containing the universal sequence to the nucleic acids, and then adding the compartment-specific index sequence to the nucleic acids. 16. A method of obtaining a library for sequencing containing nucleic acids derived from a plurality of individual nuclei or cells, where this method includes: (a) obtaining a plurality of nuclei or cells in a first plurality of compartments, where each compartment contains a subpopulation of nuclei or cells; (b) contacting each subpopulation with a reverse transcriptase and a primer to obtain double-stranded DNA nucleic acids containing the primer and the corresponding DNA nucleotide sequence from template RNA molecules; (c) processing DNA molecules in each subset of nuclei or cells to obtain indexed nuclei or cells, where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed nucleic acids present in the indexed nuclei or cells, where the processing includes ligation, primer extension, hybridization, amplification or transposition; and (d) combining indexed nuclei or cells to form combined indexed nuclei or cells; (e) distributing the pooled indexed nuclei or cells across the second set of compartments, where each compartment contains a subpopulation of nuclei or cells; (f) processing DNA molecules in each subset of nuclei or cells to obtain nuclei or cells with two indices, where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, a second compartment-specific index sequence to obtain nucleic acids with two indices present in the indexed nuclei or cells, where the processing includes ligation, primer extension, hybridization, amplification or transposition; and (g) combining nuclei or cells with two indices to obtain combined nuclei or cells with two indices; (h) the distribution of the combined nuclei or cells with two indices across the third set of compartments, where each compartment contains a subpopulation of nuclei or cells; (i) processing DNA molecules in each subset of nuclei or cells to obtain nuclei or cells with three indices, where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, a third compartment-specific index sequence to obtain nucleic acids with three indices present in the indexed nuclei or cells, where the processing includes ligation, primer extension, hybridization, amplification or transposition; and (j) combining nuclei or cells with three indices to obtain combined nuclei or cells with three indices. 17. A method of obtaining a library for sequencing containing nucleic acids derived from a plurality of individual nuclei or cells, where this method includes: (a) obtaining a plurality of nuclei or cells; (b) contacting a plurality of nuclei or cells with a reverse transcriptase and a primer to produce double-stranded DNA nucleic acids containing the primer and the corresponding DNA nucleotide sequence from the RNA template nucleic acids; (c) the distribution of nuclei or cells in the first set of compartments, where each compartment contains a subpopulation of nuclei or cells; (d) processing DNA molecules in each subset of nuclei or cells to obtain indexed nuclei or cells, where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, the first compartment-specific index sequence to obtain indexed nucleic acids present in the indexed nuclei or cells, where the processing includes ligation, primer extension, hybridization, amplification or transposition; and (e) combining indexed nuclei or cells to form combined indexed nuclei or cells; (f) distributing the pooled indexed nuclei or cells across the second set of compartments, where each compartment contains a subpopulation of nuclei or cells; (g) processing DNA molecules in each subset of nuclei or cells to obtain nuclei or cells with two indices, where the processing includes adding to the DNA nucleic acids present in each subset of nuclei or cells, a second compartment-specific index sequence to obtain nucleic acids with two indices present in the indexed nuclei or cells, where the processing includes ligation, primer extension, hybridization, amplification or transposition; and (h) combining nuclei or cells with two indices to obtain combined nuclei or cells with two indices; (i) the distribution of the combined nuclei or cells with two indices across the third set of compartments, where each compartment contains a subpopulation of nuclei or cells; (j) processing DNA molecules in each subset of nuclei or cells to obtain nuclei or cells with three indices, where the processing includes adding to the DNA nucleic acids present in subpopulations of nuclei or cells, a third compartment-specific index sequence to obtain nucleic acids with three indices present in the indexed nuclei or cells, where the processing includes ligation, primer extension, hybridization, amplification or transposition; and (k) combining nuclei or cells with three indices to obtain combined nuclei or cells with three indices. 18. The method according to claim 16 or 17, a) where the primer hybridizes with RNA nucleic acids to form double-stranded DNA nucleic acids containing the primer and the corresponding DNA nucleotide sequence from template RNA molecules, optionally i) where the primer contains a poly-T nucleotide sequence that hybridizes to the mRNA poly(A) tail, optionally, where contacting also includes contacting subpopulations with a second primer, where the second primer contains a sequence that hybridizes to a predetermined DNA nucleic acid, optionally where the second primer contains a compartment-specific index, or ii) where the primer contains a sequence that hybridizes to a predefined RNA nucleic acid, optionally where the predefined RNA nucleic acid is mRNA, or b) where the primer includes a template switch primer, or c) where the processing for adding one or more first, second, or third compartment-specific index sequences comprises a two-step method of adding a nucleotide sequence containing a universal sequence to nucleic acids, and then adding a first compartment-specific index sequence to DNA nucleic acids, or d) where the primer contains the first compartment-specific index sequence, or e) further comprising, before contacting, labeling newly synthesized RNA in subpopulations of cells or nuclei isolated from cells, optionally, i) where already existing RNA nucleic acids and newly synthesized RNA nucleic acids are labeled with the same indices in the same compartment, or ii) where the labeling comprises the incubation of a plurality of nuclei or cells in a composition containing a nucleotide tag, where the nucleotide tag is incorporated into the newly synthesized RNA, or iii) where the nucleotide tag contains a nucleotide analog, a hapten-labeled nucleotide, a mutagenic nucleotide, or a nucleotide that can be modified by a chemical reaction, or iv) where more than one nucleotide tag is included in the newly synthesized RNA, optionally where the ratios of the nucleotide tag or tags are different for different compartments or time points, or v) also comprising treating a subpopulation of nuclei or compartment cells under predetermined conditions prior to labeling, optionally a) where the predetermined condition includes treatment with an agent, b) where the predetermined conditions of the two or more compartments are different, or c) where the treatment and labeling are performed simultaneously or the treatment is prior to labeling, optionally wherein the agent comprises a protein, a non-ribosomal protein, a polyketide, an organic molecule, an inorganic molecule, an RNA or RNAi molecule, a carbohydrate, a glycoprotein, a nucleic acid, or a combination thereof, - where the agent includes a therapeutic drug, f) where one or more distributions include dilution, g) where one or more distributions involve sorting, h) wherein the addition of one or more of the first, second, or third compartment-specific index sequences comprises contacting the subpopulations with the hairpin ligation duplex under conditions suitable for ligating the hairpin ligation duplex to the end of the nucleic acid fragments, i) wherein the addition of one or more of the first, second, or third compartment-specific index sequences comprises contacting the nucleic acid fragments with a transposome complex, where the transposome complex in the compartments contains a transposase and a universal sequence, where the contacting also includes conditions suitable for fragmentation of the nucleic acid fragments and incorporating the nucleotide sequence into nucleic acid fragments, j) where adding a first or second compartment-specific index involves ligation and adding a subsequent compartment-specific index sequence involves transposition, k) where the compartment includes a well or a drop, l) where the compartments of the first set of compartments contain from 50 to 100,000,000 nuclei or cells, m) where the compartments of the second set of compartments contain from 50 to 100,000,000 nuclei or cells, n) where the compartments of the third set of compartments contain from 50 to 100,000,000 nuclei or cells, o) further comprising obtaining nucleic acids with three indexes from pooled nuclei or cells with three indexes, and thereby obtaining a library for sequencing from a plurality of nuclei or cells, p) further comprising: obtaining a surface containing a plurality of amplification sites, where the sites of amplification include at least two populations of linked single-stranded capture oligonucleotides having a free 3' end, and contacting a surface containing amplification sites with nucleic acid fragments with three indexes under conditions suitable for obtaining a plurality of amplification sites, each containing a clonal population of amplicons, from a single fragment containing a plurality of indices. 19. A method for obtaining a sequencing library containing nucleic acids from a plurality of single cells, wherein said method includes: (a) obtaining nuclei from a plurality of cells; (b) distributing nuclear subpopulations across a first plurality of compartments and contacting each subpopulation with a reverse transcriptase and a primer, where the primer in each compartment contains a first index sequence that differs from first index sequences in other compartments, to obtain indexed nuclei containing indexed nucleic acid fragments ; (c) combining indexed kernels to obtain combined indexed kernels; (d) allocating subpopulations of the pooled indexed nuclei to a second plurality of compartments and contacting each subpopulation with the hairpin ligation duplex under conditions suitable for ligating the hairpin ligation duplex to the end of the indexed nucleic acid fragments containing the first index sequence to produce dual-index nuclei containing the fragments a nucleic acid with two indexes, where the hairpin ligation duplex contains a second index sequence different from the second index sequences in other compartments; (e) merging the 2-index kernels to produce merged 2-index kernels; (f) distributing the subpopulations of the dual-index pooled nuclei across a third set of compartments and processing the dual-index nucleic acid fragments under conditions suitable for second strand synthesis; (g) contacting the two-index nucleic acid fragments with a transposome complex, wherein the transposome complex in each compartment contains a transposase and a universal sequence, where the contacting includes conditions suitable for fragmenting the two-index nucleic acid fragments and incorporating the universal sequence into the two-index nucleic acid fragments. indices to obtain nucleic acid fragments with two indices, containing the first and second indices at one end and the universal sequence at the other end; (i) inserting into nucleic acid fragments with two indexes in each compartment a third index sequence to obtain fragments with three indexes; (j) combining tri-index fragments to obtain a sequencing library containing transcriptome nucleic acids from a plurality of individual cells. 20. The method according to p. 19, where the primer contains a poly-T sequence that hybridizes to the poly(A) tail of the mRNA, or - where the primer of each compartment contains a sequence that hybridizes to a predefined mRNA, optionally, where the method includes primers in different compartments that hybridize to different nucleotides of the same predefined mRNA. 21. A method for obtaining a library for sequencing a transcriptome containing nucleic acids from a plurality of individual cells, where this method includes: (a) obtaining united nuclei from a plurality of cells; (b) contacting the pooled nuclei with reverse transcriptase and a primer containing the oligo-dT sequence, which hybridizes to the mRNA poly(A) tail to form pooled nuclei containing nucleic acid fragments; (c) distributing subpopulations of pooled nuclei across a plurality of compartments and contacting each subpopulation with a hairpin ligation duplex under conditions suitable for ligating the hairpin ligation duplex to the end of the nucleic acid fragments to produce indexed nuclei containing indexed nucleic acid fragments, where the hairpin ligation duplex contains an index sequence , different from index sequences in other compartments; (d) combining indexed kernels to obtain combined indexed kernels; (e) allocating subpopulations of the pooled indexed nuclei to a second set of compartments and processing the indexed nucleic acid fragments under conditions suitable for second strand synthesis; (f) contacting the indexed nucleic acid fragments with a transposome complex, where the transposome complex in each compartment contains a transposase and a universal sequence, where the contacting includes conditions suitable for fragmenting the indexed nucleic acid fragments and incorporating the universal sequence into the indexed nucleic acid fragments to obtain indexed nucleic acid fragments acids containing an index at one end and a universal sequence at the other end; (g) incorporating into the indexed nucleic acid fragments in each compartment a second index sequence to produce fragments with two indexes; (h) combining fragments with two indexes to obtain a sequencing library containing transcriptome nucleic acids from a plurality of individual cells.

Images