Review

Detection of mRNA Transcript Variants

Kevin Vo, Sharmin Shila , Yashica Sharma, Grace J. Pei, Cinthia Y. Rosales, Vinesh Dahiya , Patrick E. Fields
and M. A. Karim Rumi *

Department of Pathology and Laboratory Medicine, University of Kansas Medical Center,

Kansas City, KS 66160, USA; kvo5@kumc.edu (K.V.); sharminshila.mib@gmail.com (S.S.);
yashica2025@gmail.com (Y.S.); 3096823@smsd.org (G.J.P.); 255cdy30@students.olatheschools.com (C.Y.R.);
vinesh.dahiyampharm@gmail.com (V.D.); pfields@kumc.edu (P.E.F.)
* Correspondence: mrumi@kumc.edu; Tel.: +1-(913)-588-8059

Abstract: Most eukaryotic genes express more than one mature mRNA, defined as tran-
script variants. This complex phenomenon arises from various mechanisms, such as using
alternative transcription start sites and alternative post-transcriptional processing events.
The resulting transcript variants can lead to synthesizing proteins that possess distinct func-
tional domains or may even generate noncoding RNAs, each with unique roles in cellular
processes. The generation of these transcript variants is not merely a random occurrence; it
is cell-type specific and varies with developmental stages, aging processes, or pathogenesis
of diseases. This highlights the biological significance of transcript variants in regulating
gene expression and their potential impact on cellular functionality. Despite the biological
importance, investigating transcript variants has been hampered by challenges associated
with detecting their expression. This review article addresses the advancements in molecu-
lar techniques in detecting transcript variants. Traditional methods such as RT-PCR and
RT-qPCR can easily detect known transcript variants using primers that target unique exons
associated with the variants. Other techniques like RACE-PCR and hybridization-based
methods, including Northern blotting, RNase protection assays, and microarrays, have
also been utilized to detect transcript variants. Nevertheless, RNA sequencing (RNA-Seq)
has emerged as a powerful technique for identifying transcript variants, especially those
with previously unknown sequences. The effectiveness of RNA sequencing in transcript
variant detection depends on the specific sequencing approach and the precision of data
analysis. By understanding the strengths and weaknesses of each laboratory technique,
Academic Editor: Christiane Branlant researchers can develop more effective strategies for detecting mRNA transcript variants.
Received: 18 January 2025 This ability will be crucial for our comprehensive understanding of gene regulation and
Revised: 13 March 2025 the implications of transcript diversity in various biological contexts.
Accepted: 15 March 2025
Published: 16 March 2025
Keywords: gene expression; mRNA transcript variants; RNA sequencing; spatial
Citation: Vo, K.; Shila, S.; Sharma, Y.; transcriptomics; analyses of NGS data; verification of NGS data
Pei, G.J.; Rosales, C.Y.; Dahiya, V.;
Fields, P.E.; Rumi, M.A.K. Detection of
mRNA Transcript Variants. Genes 2025,
16, 343. https://doi.org/10.3390/
genes16030343
1. Introduction
Mammalian cells are versatile and complex in expressing various types of ribonucleic
Copyright: © 2025 by the authors.
Licensee MDPI, Basel, Switzerland.
acids (RNAs), each transcribed by distinct RNA polymerases that play essential roles
This article is an open access article in cellular function. Specifically, RNA polymerase I (Pol I) transcribes ribosomal RNAs,
distributed under the terms and which are fundamental components of the ribosome and necessary for protein synthesis.
conditions of the Creative Commons RNA polymerase II (Pol II), on the other hand, transcribes precursor messenger RNAs
Attribution (CC BY) license
(pre-mRNAs), which serve as templates for protein-coding sequences. Meanwhile, RNA
(https://creativecommons.org/
polymerase III (Pol III) transcribes transfer RNAs (tRNAs) and a variety of other small RNAs
licenses/by/4.0/).

Genes 2025, 16, 343 https://doi.org/10.3390/genes16030343

Genes 2025, 16, 343 2 of 18

that are crucial for various cellular processes [1]. While a subset of pre-mRNAs is processed
into mature protein-coding mRNAs, a significant proportion of the transcripts generated by
RNA polymerases I, II, and III are classified as noncoding RNAs. These noncoding RNAs
play critical roles in regulating gene expression at multiple levels, including transcriptional
regulation, post-transcriptional processing of RNAs, and the translation of mRNAs into
proteins [2,3]. The noncoding RNAs include ribosomal RNA (rRNA), long noncoding RNA
(lncRNA), small nuclear RNA (snRNA), transfer RNA (tRNA), circular RNA (circRNA),
and micro-RNA (miRNA) [3]. Although the protein-coding pre-mRNAs, as well as the
noncoding RNAs, undergo post-transcriptional processing to generate mature transcripts,
the primary focus of this article is the detection of mRNA transcript variants.
The mammalian genome contains an estimated 20,000–30,000 protein-coding genes [4].
However, 95% of human genes undergo alternative splicing, producing an average of three
mature mRNA variants per gene [5]. In addition, alternative transcription start sites (TSSs)
and alternative transcription termination/polyadenylation (APA) increase the diversity
in mRNA transcript variants [6]. As a result, the number of different mature mRNAs is
estimated to be 60,000–90,000, outnumbering the genes from which they are derived [6,7].
Different proteins can be encoded by different transcript variants. The number of proteins
that potentially can be encoded by mRNA transcripts is significantly higher, estimated to
be in the range of 80,000 to 120,000 [8,9].
Gene expression studies remain fundamental to understanding gene regulation and
cellular functions [10]. Despite their potential biological significance, the functional roles of
transcript variants derived from gene expression analyses are often overlooked. Disregard-
ing the analyses of transcript variants can pose a significant limitation in quantifying their
expression, which is essential for gaining information about complex biological processes.
The expression profiles of such variants give a measure of protein variations and protein
expression, allowing for the studies of protein functions [9].

2. Importance of Detecting mRNA Transcript Variants

A common misconception in gene expression analyses is the oversimplified notion
that each gene corresponds to a single mRNA transcript, which, in turn, is used to produce
a single protein [11]. This perspective is reductive and biologically inaccurate, especially
in the context of mammalian genes that can generate multiple transcript variants [12].
The primary mechanisms responsible for increasing the diversity in mRNAs include the
initiation of transcription from alternative start sites, alternative splicing, and APA [6].
These variants can differ significantly in several aspects, including the structure of their
5′ -ends, the composition of their exons, and their 3′ -ends [13–15]. Alternative start sites
increase variability at the 5′ -end of the transcript and may encode proteins with diverse
amino termini, and APA alters the 3′ -end of mRNA and generates protein diversity at the
carboxy termini [16,17]. However, alternative splicing can impact any part of the coding
sequences. For example, the mouse Runx1 gene expresses transcript variants with different
TSSs, APA sites, and alternative splicing [18] (Figure 1). It is also well-known that the pre-
mRNAs expressed from different members of the human BCL2 family undergo alternative
splicing to form multiple isoforms [19]. While BCL2 forms BCL-2α and BCL-2β, BCL-X
(BCL2L1) forms BCL-XL and BCL-XS , and MCL-1 forms MCL-1L, MCL-1S, and MCL-1ES.
In addition, both BCL-W (BCL2L2) and BFL-1/A1 form two isoforms due to alternative
splicing [19,20].
 noncoding RNAs [23]. Thus, some variants may possess a more critical role than other
transcript variants. The traditionally focused transcript variant approach overlooks the
expression and function of remaining transcript variants. Accordingly, detecting all the
Genes 2025, 16, 343 transcript variants remains fundamental to understanding cell-type-specific gene regula-
3 of 18
tion and cellular phenotypes.

Figure
Figure 1. ExpressionofofmRNA
1. Expression mRNA transcript
transcript variants.
variants. A schematic
A schematic diagram
diagram showing
showing the transcript
the transcript var-
variants of mouse Runx1. The transcript variants were expressed due to alternative
iants of mouse Runx1. The transcript variants were expressed due to alternative transcription transcription
start
start sites (TSSs)
sites (TSSs) in all
in all the the variants
variants (201 to(201
207),to 207), alternative
alternative splicingsplicing
(variant (variant
202), and202), and alternative
alternative polyad-
polyadenylation sites (variants 202, 203, 204, and 206). This figure has been adapted from the mouse
enylation sites (variants 202, 203, 204, and 206). This figure has been adapted from the mouse Runx1
Runx1 transcript variant map on the ENSEMBL website (not to scale).
transcript variant map on the ENSEMBL website (not to scale).
While some transcript variants expressed from a gene can translate into different
The generation of transcript variants can be cell-type specific and linked to develop-
proteins, others may form noncoding regulatory RNAs. Although the protein-coding open
mental stages or disease conditions [24]. The same gene may express different transcript
reading frames (ORFs) often remain intact in transcript variants expressed by the same
variants in different tissues. The same cell lineage may also express different transcript
gene, proteins with different amino acid sequences can also be encoded by variants that
variants from a single gene at different differentiation or developmental stages. Therefore,
gain a frameshift or isoform-specific unique sequences [21,22]. Mature mRNAs that lose
analyzing transcript switching is crucial for understanding cell differentiation and cell fate
their ORFs and are not translated into functional proteins may undergo decay or act as
determination. Moreover, mutations or disease conditions may lead to the expression of
noncoding RNAs [23]. Thus, some variants may possess a more critical role than other
different transcript variants from a single gene in the same cell type. Recent studies have
transcript variants. The traditionally focused transcript variant approach overlooks the
suggested that altered mRNA transcript variants may be involved in disease pathogene-
expression and function of remaining transcript variants. Accordingly, detecting all the
sis, including carcinogenesis [25]. Thus, identifying the disease-specific transcript variants
transcript variants remains fundamental to understanding cell-type-specific gene regulation
may serve as biomarkers for disease diagnosis and provide a potential target for drug
and cellular phenotypes.
delivery or therapeutic measures. It has been suggested that mutations may impact pre-
The generation of transcript variants can be cell-type specific and linked to develop-
mRNA splicing and cause diseases [26]. For example, hypercholesterolemia may result
mental stages or disease conditions [24]. The same gene may express different transcript
from mutated exon sequences of LDL receptors caused by the dysregulation of alternative
variants in different tissues. The same cell lineage may also express different transcript
splicing [27]. Although alternative TSSs, alternative splicing, and APA allow for the for-
variants from a single gene at different differentiation or developmental stages. Therefore,
mation of functionally diverse transcript variants, they have also been reported in differ-
analyzing transcript switching is crucial for understanding cell differentiation and cell fate
ent types of cancers; detecting these variants and examining the underlying mechanisms
determination. Moreover, mutations or disease conditions may lead to the expression of
are emerging fields of cancer biology and should become focal points to the challenges of
different transcript variants from a single gene in the same cell type. Recent studies have
cancer prevention [28].
suggested that altered mRNA transcript variants may be involved in disease pathogenesis,
including carcinogenesis [25]. Thus, identifying the disease-specific transcript variants
3. Detection of Transcript Variants
may serve as biomarkers for disease diagnosis and provide a potential target for drug
Transcript
delivery variants measures.
or therapeutic expressed by different
It has been genes werethat
suggested identified long may
mutations before genome-
impact pre-
wide approaches were developed [29,30]. Cloning and sequencing of mRNA
mRNA splicing and cause diseases [26]. For example, hypercholesterolemia may result libraries,
Northern
from blotting,
mutated exon microarrays,
sequences of RNase protection
LDL receptors assays,
caused RACE-PCR,
by the ddPCR,
dysregulation and RT-
of alternative
splicing [27]. Although alternative TSSs, alternative splicing, and APA allow for the forma-
tion of functionally diverse transcript variants, they have also been reported in different
types of cancers; detecting these variants and examining the underlying mechanisms are
emerging fields of cancer biology and should become focal points to the challenges of
cancer prevention [28].

3. Detection of Transcript Variants

Transcript variants expressed by different genes were identified long before genome-
wide approaches were developed [29,30]. Cloning and sequencing of mRNA libraries,
Genes 2025, 16, 343 4 of 18

Northern blotting, microarrays, RNase protection assays, RACE-PCR, ddPCR, and RT-
PCR have made notable contributions in identifying transcript variants. However, these
techniques are suitable for selected genes, not whole transcriptome studies [31]. Recent
advancements in RNA sequencing (RNA-Seq) techniques have allowed the scientific com-
munity to identify transcript variants and analyze their development mechanism and
potential role in cellular functions. In the following sections, we discuss RNA-Seq in
detail, followed by the abovementioned techniques that can be used to verify selected
transcript variants.

3.1. RNA Sequencing

RNA-Seq can detect and analyze the sequences of RNA molecules present in a
test sample [32]. This technique can elucidate the complexity of transcription and post-
transcriptional processing of pre-mRNAs that form mature mRNAs. The first step of
RNA-Seq is assessing the RNA quality using an Agilent Bioanalyzer or similar techniques.
Then, high-quality total RNA (e.g., RIN value ≥ 8) or purified mRNA are reverse tran-
scribed using oligo(dT) primers [33]. RNA-seq can also be performed with random primers,
especially for transcript detection with short-read methods. Purified mRNAs are chem-
ically fragmented for short-read sequencing, and then cDNAs are prepared. After the
preparation of cDNA libraries (single- or double-stranded), adapters are ligated to either
the intact (for long-read sequencing) or fragmented cDNAs (for short-read sequencing).
Then, the libraries can be sequenced on strategy-specific platforms directly or after PCR
amplifications (Figure 2).
The results are demultiplexed according to test samples and processed for alignment,
assembly, and further analysis. In an RNA-seq experiment, identification of the alternative
TSSs, alternative splicing, and APA depends on the quality of RNA, library preparation,
sequencing platform, and data analysis [34]. Transcript variant analysis can be improved
with paired-end sequencing due to the heavy constraints on the distance between reads
when mapping along the reference sequence [35]. The counted reads aligning with each
transcript quantify the transcript expression normalized by the transcript length. This
process of paired-end sequencing allows for better differentiation amongst currently known
variants and a total quantification of reads [35].
Recent advances in RNA-Seq technology have expanded its applications in various
experimental settings. Illumina (San Diego, CA, USA), Oxford Nanopore Technologies
(ONT, Oxford, UK), and Pacific Biosciences (PacBio, Menlo Park, CA, USA) are the most
popular RNA-Seq methods [36]. ONT and PacBio can perform long-read sequencing;
however, ONT uniquely enables the direct sequencing of full-length RNA molecules
without converting them to cDNA. This allows ONT to detect RNA modifications, such as
methylation, alongside nucleotide sequences [37]. In contrast, PacBio relies on cDNA-based
methods for RNA sequencing and cannot directly sequence native RNA molecules [38].
Other sequencing methods include single-cell RNA sequencing (scRNA-Seq) using 10x
Genomics (Pleasanton, CA, USA), Takara Bio (SMART-Seq) (San Jose, CA, USA), or Bio-
Rad (ddSEQ, Hercules, CA, USA) systems to render various cell libraries at once and
view the cellular landscape of tissues. The scRNA-seq libraries are run on a standard
sequencing platform (Illumina, ONT, or PacBio), and the sequencing data are analyzed
using specific software.
Bulk RNA-Seq (e.g., Illumina, ONT, and PacBio) uses RNA samples to provide gene
expression profiles across tissue or a population of cells that may contain more than one
cell type [39]. Despite the cost-effectiveness of bulk sequencing, it represents a mixed
expression profile of all the cell types present in the tissue. This process fails to distinguish
genes expressed in specific cell types and captures cellular heterogeneity [40]. In contrast,
Genes 2025, 16, 343 5 of 18

Genes 2025, 16, x FOR PEER REVIEW 5 of 19

scRNA-Seq (e.g., 10x Genomics) allows for the detection of cellular heterogeneity and
identification of individual cell populations while also giving gene expression profiles for
the libraries
the libraries it
it creates
creates [41,42].
[41,42]. These
These expression
expression profiles
profiles can
can bebe analyzed
analyzed viavia aa standard
standard
RNA-seq platform. However, the gene expression data at the single-cell
RNA-seq platform. However, the gene expression data at the single-cell level come withlevel come witha
a few limitations. In addition to the higher cost and computational complexity,
few limitations. In addition to the higher cost and computational complexity, obtaining obtaining
fresh cells
fresh cellsby
byremoving
removingdead deadcells
cellsand
andcell
celldebris
debris remains
remains a limitation
a limitation to generating
to generating pre-
precise
cise transcriptome
transcriptome data data [40,43].
[40,43].

Figure2.2.Overview
Figure Overviewofofshort-read
short-readmRNA
mRNAsequencing.
sequencing.A
A schematic
schematic presentation
presentation of
of stranded
stranded mRNA
mRNA
library preparation using an Illumina kit. The process begins with mRNA enrichment via
library preparation using an Illumina kit. The process begins with mRNA enrichment via poly(A) poly(A)
selection and fragmentation. This is followed by first- and second-strand cDNA synthesis. The
selection and fragmentation. This is followed by first- and second-strand cDNA synthesis. The re-
resulting double-stranded cDNA undergoes end repair and adenylation before ligating indexed
sulting double-stranded cDNA undergoes end repair and adenylation before ligating indexed
adapters. The library is then amplified and cleaned. The final product is an indexed library ready for
adapters. The library is then amplified and cleaned. The final product is an indexed library ready
Illumina sequencing, containing elements like P5/P7 sequences and sequencing primer binding sites.
for Illumina sequencing, containing elements like P5/P7 sequences and sequencing primer binding
3.1.1.
sites. Short-Read Versus Long-Read mRNA Sequencing
Short-read mRNA sequencing (e.g., Illumina) involves isolating poly-A mRNA using
3.1.1. Short-Read
oligo(dT) magneticVersus Long-Read
beads, followed mRNA
by the Sequencing
fragmentation of mRNAs and reverse tran-
Short-read
scription mRNA sequencing
into first-strand cDNA with(e.g., Illumina)
random involves
primers. isolating
Strand poly-AismRNA
specificity using
achieved by
oligo(dT) magnetic
incorporating dUTP beads, followed
in the second by the
strand, fragmentation
which of mRNAs
is later degraded. and are
Adapters reverse tran-
ligated to
scription into first-strand cDNA with random primers. Strand specificity is achieved by
incorporating dUTP in the second strand, which is later degraded. Adapters are ligated
Genes 2025, 16, 343 6 of 18

the fragments, and the library is amplified via PCR (Figure 2). Sequencing can be performed
using Illumina’s sequencing-by-synthesis (SBS) technology, offering accurate transcriptome
analysis [44,45].
In contrast to short-read mRNA sequencing, long-read mRNA is a powerful tech-
nique that allows reading the full-length RNA molecules without fragmentation [46]. This
method identifies the sequences of full-length mRNA transcripts, which can be analyzed
to determine alternative TSSs, splicing events, APA, gene mutations, and gene fusions
accurately [47]. ONT offers two methods for long-read mRNA sequencing: direct RNA
and cDNA-based. Direct RNA sequencing isolates poly-A RNA using poly-T oligo beads,
ligates adapters directly to native RNA without reverse transcription, and sequences it
through nanopores, preserving RNA modifications and strand specificity [48,49]. Although
ONT’s direct RNA sequencing takes steps to avoid most biases, the poly-A selected RNA
populations introduce a capture bias that could be misinterpreted as differential gene
expression [50]. This 3′ bias occurs when the poly-A selection skews the sequenced mRNAs
toward the lengthier tails, creating a distorted view of the transcriptome. It is best to omit
the selection during pre-processing when sequencing with a poly-A selection RNA popula-
tion [50]. On the other hand, cDNA-based sequencing involves the reverse transcription
of poly-A RNA into full-length cDNA, followed by adapter ligation and amplification,
providing higher throughput but losing RNA modification information [32].
PacBio follows a similar cDNA-based approach to sequencing with the current Iso-
Seq technique where cDNA is synthesized; random primers are annealed for first-strand
synthesis, and reverse transcription activates template switching, followed by cDNA
amplification and adapter ligation. These sequenced full-length transcript reads have a
maximum insertion size of 10 kilobase pairs [51]. Circular consensus reads, such as HiFi
reads, can improve accuracy for isoform detection and transcriptome annotation when
combined in the same pipeline with Iso-Seq. Overall, short-read sequencing is typically
highly accurate and ideal for gene expression profiling and SNP detection but struggles
to resolve full-length transcripts or complex genomic regions due to its short read lengths
(50–300 bp) [52,53]. Long-read sequencing captures full-length transcripts, resolves struc-
tural variants, and detects isoforms.

3.1.2. Direct Versus PCR-Amplified Detection of mRNAs

Direct mRNA sequencing (dRNA-Seq) allows the direct detection of full-length mRNA
transcripts and the characterization of RNA modifications [48]. After purifying the mRNAs
from total RNAs and assessing their quality, dRNA-Seq can be performed using ONT
technology. Libraries are prepared using oligo(dT) primers, and first-strand cDNA is
synthesized by reverse transcription [54]. Then, the mRNA-cDNA hybrid is purified, and
the sequencing adapters are ligated, purified again, and loaded onto platform-specific
Flow Cells to run for sequencing. The protocol for cap-dependent ligation includes de-
phosphorylating and de-capping mRNA and ligating a biotinylated 5′ adapter RNA [55].
After purifying and reassessing, processing is carried out with the library preparation, as
described [56]. This method provides comprehensive insights into RNA molecules in their
native form [48].
PCR-amplified RNA sequencing involves increasing the quantity of target cDNA li-
braries using 5 to 10 cycles of PCR. The PCR primers can bind to the adapters ligated to
cDNA ends. The cDNA library is amplified to generate adequate cDNA material for sequenc-
ing. After purifying the PCR-amplified cDNA library, the final step is to sequence on an
appropriate platform [57]. There are some potential drawbacks to the PCR amplification of
RNA-Seq libraries [32]. PCR amplification of the cDNA sequences is not linear and introduces
amplification bias due to primer bias, GC content bias, and secondary structure bias [58,59].
Genes 2025, 16, 343 7 of 18

Amplification bias may also result in over-representation of the abundant transcripts and
sequencing errors due to base misincorporation [60,61]. Recent studies have suggested in-
corporating unique molecular identifiers (UMIs), the optimization of PCR conditions, and
low-cycle PCR to mitigate PCR amplification bias [61–63].

3.1.3. Bulk Sequencing Versus Single-Cell Sequencing of mRNAs

Bulk RNA-seq is a method for analyzing the whole transcriptome of a sample by
sequencing the mRNAs irrespective of cellular origin. Bulk sequencing is performed using
either a short-read or long-read strategy. After converting the intact or fragmented mRNAs
to cDNA, sequencing adapters are ligated. Then, the cDNA libraries are processed with or
without PCR amplification and run on a specific sequencing platform. The sequencing data
output is demultiplexed according to the adapter sequences and analyzed using different
commercial or open software.
Single-cell mRNA sequencing can be performed using flow sorting (Takara Bio USA,
San Jose, CA, USA) or microfluidics methods (10x Genomics). Takara Bio’s SMART-Seq
mRNA Single-Cell LP protocol is optimized for generating high-quality Illumina-ready
libraries from single cells, especially those with a low RNA content. It uses SMART (Switch-
ing Mechanism at the 5′ -end of RNA Template) technology, which employs a template-
switching reverse transcriptase to enrich full-length cDNA and add PCR adapters directly to
both ends. This ligation-free workflow minimizes handling errors and accurately represents
mRNA transcripts, including the 5′ -ends. The protocol involves enzymatic fragmentation,
stem–loop adapter ligation, and library amplification, producing sequencing-ready libraries
within two days [64]. For 10x Genomics, cells move through channels within Chromium
X instruments at a limited dilution and generate nanometer-sized gel beads-in-emulsion
(GEMs). This technique uses microfluidic partitioning to encapsulate single cells in GEMs
containing barcoded oligonucleotides. Reverse transcription occurs within GEMs, attaching
UMIs and cell barcodes to cDNA. After breaking the emulsion, cDNA is amplified, frag-
mented, and prepared into sequencing-ready libraries compatible with Illumina platforms.
This method enables the high-throughput processing of thousands of cells in a single run,
capturing 3′ gene expression profiles with high resolution. Data analysis is performed
using the Cell Ranger pipeline for cell-specific transcriptome mapping [65,66].
Single-cell mRNA sequencing, such as Takara Bio’s SMART-Seq and 10x Genomics,
captures cellular heterogeneity and rare populations and is valuable for studying cell
types in complex tissues, such as the brain or immune system, where understanding
cellular differences is crucial [67]. Takara excels in full-length transcript analysis using
SMART technology, while 10x enables high-throughput profiling of thousands of cells
via microfluidic barcoding methods [68,69]. Consequentially, due to the low capture
efficiency of isoforms by scRNA-Seq overall, forced detection of splicing events results in
confounding [70]. Since the amount of single-cell material initiating library preparation is
too small, the frequency of dropouts in splicing analysis becomes apparent. This limitation
will remain unless the preparation for scRNA-Seq changes, which poses a significant
challenge. Hence, it is recommended not to attempt large-scale alternative splicing analysis
during scRNA-Seq. Bulk RNA-seq, by contrast, measures average gene expression across
cell populations, making it cost-effective for global transcriptome analysis but unable to
resolve individual cell types or rare populations [40,71].

3.1.4. Analysis of RNA Sequencing Data

Data analysis begins with importing the RNA-Seq data to the pipeline and removing
the adapter sequences from cDNA libraries [72]. For the short-read sequence analysis,
each cDNA fragment sequence is read through a commercial data-processing platform,
Genes 2025, 16, 343 8 of 18

Genes 2025, 16, x FOR PEER REVIEW 8 of 19

such as CLC Genomics (Qiagen Bioinformatic), Partek Flow (Illumina), or DNASTAR

Lasergene (DNASTAR, Madison, WI, USA). After reading the nucleotide sequences of
Lasergene (DNASTAR, Madison, WI, USA). After reading the nucleotide sequences of the
the cDNA fragments, the fragments are assembled and aligned to a reference genome
cDNA fragments, the fragments are assembled and aligned to a reference genome to en-
to ensure accurate mapping of each read to the correct location. These aligned reads are
sure accurate mapping of each read to the correct location. These aligned reads are quan-
quantified to determine gene expression levels, involving counting and normalizing the
tified to determine gene expression levels, involving counting and normalizing the reads
reads corresponding to each gene [73]. Data alignment against a reference genome identifies
corresponding to each gene [73]. Data alignment against a reference genome identifies the
the splice junctions and denotes the alternatively spliced variants [74].
splice junctions and denotes the alternatively spliced variants [74].
After such alignments, gene and transcript level quantifications are undertaken by
After such alignments, gene and transcript level quantifications are undertaken by
normalizing
normalizing with withthe
thetotal
totalread
readcounts
counts (e.g.,
(e.g., TPM,
TPM, RPKM,
RPKM, etc.).
etc.). Gene-level
Gene-level quantification
quantification
assigns all the cDNA fragments to a gene, where all transcripts
assigns all the cDNA fragments to a gene, where all transcripts are aligned to that are aligned to that
genegene
locus [75]. Generally,
locus [75]. Generally,the thegene
geneexpression
expression (GE)
(GE) values
values represent
represent thethe
sumsum ofexpressions
of all all expressions
from the transcript variants expressed from that gene [76,77]. Transcript-level
from the transcript variants expressed from that gene [76,77]. Transcript-level quantifica- quantification
assigns cDNA
tion assigns cDNAfragments
fragments to the reference
to the reference transcript sequences.
transcript sequences. Although
Althoughthe thefragments
frag-
may be ambiguous, alignment with the reference transcripts allows
ments may be ambiguous, alignment with the reference transcripts allows for superior for superior biological
resolution, giving insight
biological resolution, givinginto isoform
insight into switching, which iswhich
isoform switching, overlooked during during
is overlooked gene-level
quantification [78]. Once[78].
gene-level quantification the Once
assembly, alignment,
the assembly, and quantification
alignment, and quantificationare completed,
are com- dif-
pleted, differential
ferential expression expression
analysis analysis can be employed
can be employed to understand
to understand the regulation
the regulation of or
of gene
gene or transcript
transcript expression.
expression. Short-read
Short-read RNA-Seq RNA-Seq data
data face face difficulty
difficulty identifying
identifying tran-vari-
transcript
scriptdue
ants variants duelimited
to their to theirread
limited read These
length. length.include
These include difficulties
difficulties in mapping
in mapping readsreads
uniquely
uniquely
to specifictoisoforms,
specific resolving
isoforms, complex
resolvingsplicing
complexevents,
splicingand events,
handling and multi-mapped
handling multi- reads
mapped reads in repetitive or homologous regions and biases in
in repetitive or homologous regions and biases in quantification methods like TPM or quantification methods
like TPM
RPKM or may
that RPKM that
not may not
entirely entirely
correct forcorrect for sequencing
sequencing artifacts(Figure
artifacts [79,80] [79,80] 3).
(Figure 3).

Figure 3. Comparisonofofshort-read
3. Comparison short-readand and long-read
long-read RNA-seq
RNA-seq data
data analysis.
analysis. A schematic
A schematic presentation
presentation
of Runx1 transcript assembly and variant detection using short-read (A) and long-read
of Runx1 transcript assembly and variant detection using short-read (A) and long-read (B) sequenc- (B) sequenc-
ing. Short-read sequencing requires assembly, leading to an inefficient and potentially
ing. Short-read sequencing requires assembly, leading to an inefficient and potentially inaccurate inaccurate
view of the
view of the isoform
isoformrepertoire.
repertoire.InIncontrast,
contrast, long-read
long-read sequencing,
sequencing, withwith its ability
its ability to sequence
to sequence full- full-
length transcripts, eliminates the need for assembly and provides a complete view of the Runx1
isoform repertoire.
Genes 2025, 16, 343 9 of 18

Unlike their long-read counterparts, more pipelines and open software are available for
short reads, such as the Illumina data set. However, long-read sequencing technologies can
still enable precise identification of transcript isoforms by directly sequencing full-length
transcripts and capturing exon–intron structures and alternative splicing patterns without
assembly. These methods provide detailed isoform-specific quantification, overcoming
the limitations of short-read sequencing [81]. Challenges such as higher raw error rates
necessitate robust error correction tools like Minimap2, assembled by Canu or Miniasm,
and Iso-Seq pipelines. These ensure accuracy while addressing small exon alignment and
splice site prediction [51].

3.2. Hybridization-Based Techniques

3.2.1. Spatial Transcriptomics
Spatial transcriptomics is an advanced technique that allows researchers to map gene
expression within the context of tissue architecture [82,83]. It begins with the preparation
of tissue sections to maintain their structure [84]. Specific probes are introduced to target
mRNA molecules within the tissue. The probes that bind to target mRNAs are detected
using imaging or sequencing techniques [85]. This enables the detection of RNA, aiding in
identifying cell types based on their gene expression [86]. The in situ sequencing method
uses padlock probes and rolling circle amplification to target and amplify specific RNA
molecules within preserved tissue sections [87,88]. The SCRINSHOT approach hybridizes
padlock probes on mRNA, followed by circularization and rolling circle amplification [89].
This method allows for the multiplexed detection of thousands of cells in tissue sections,
providing a detailed map of cell states and gene expression [89]. Despite being a powerful
tool for understanding cellular heterogeneity and spatial organization in tissues, not all
spatial transcriptomics can achieve high cellular resolution. Overall limitations in the
sequencing-based approach resolution come from the physical size of the capturing spot
since multiple cells need to be captured [90]. Imaged-based methods are more suitable for
single-cell or subcellular resolution but have the sole limitation of their optical diffraction
limit [91,92]. Both spatial transcriptomic techniques can be used for their specific strengths,
but efforts to refine their sensitivity and cellular resolution are ongoing. As spatial tran-
scriptomics evolves, the ability to visualize RNA molecules in their home environment
will allow for the accurate identification of novel transcript variants in a cell type within a
tissue section [93].

3.2.2. Microarrays
Microarrays, also known as gene chips, are powerful tools for simultaneously studying
gene expression patterns across a vast number of genes [94]. The process begins with the
extraction of mRNA from the sample of interest. This mRNA is then converted to cDNA and
labeled with fluorescent markers. The labeled cDNAs are hybridized to DNA probes on the
microarray chips, which contain thousands of probes designed to hybridize with specific
mRNA molecules. After hybridization, the chip is scanned to detect fluorescent signals [95].
These signals indicate the presence and abundance of specific mRNA molecules in the
test sample. Depending on the probe sequences, mRNA transcript variants expressed in
a particular sample can be detected [95]. The detection of transcript variants will depend
on the probe design; if the target exon is absent in a variant, it will remain undetected.
Polymorphisms or mutations also remain undetected.

3.2.3. Northern Blotting

Northern blotting is the standard method for detecting RNA expressions in particular
tissues or cell types [96]. Denatured total RNA is separated on agarose gels by electrophore-
sis and transferred to nitrocellulose or nylon membranes by the capillary method [97]. After
Genes 2025, 16, 343 10 of 18

cross-linking the RNA to the membranes by UV exposure, the membranes are hybridized
to radioactive or non-radioactively labeled RNA or DNA probes [96]. Signals from the
bound probes can be imaged by autoradiography or automated imaging systems [97].
Signals from bands with different molecular weight sizes indicate the presence of multiple
mRNA transcript variants. However, Northern blotting suffers from several limitations.
The detection of the transcript variants depends on the probe targets. The resolution of
Northern blotting is not high enough to differentiate transcripts with similar lengths [98].
Moreover, Northern blots can detect only a limited number of genes and cannot be used
for the detection of novel genes.

3.2.4. RNase Protection Assays

The RNase protection assay is a sensitive technique that aims to identify and measure
the abundance of specific mRNAs [99]. This technique utilizes a procedure that hybridizes
RNA with a radioactively labeled RNA probe designed for a particular mRNA [100]. To
design this probe, a specific DNA template is needed for in vitro transcription to synthesize
an antisense RNA [101]. During the synthesis, one of the four nucleotides is replaced with
a corresponding radioactively labeled nucleotide to allow for the identification of the probe
during later steps. After the antisense RNA probe is synthesized, the DNA template is
removed through DNase digestion. Once the probe is purified, the isolated RNA is mixed
with the labeled probe for hybridization. Then, the RNA–prehybridization mix is treated
with RNase A/RNase TI and purified with a phenol–chloroform mix. While hybridization
protects the targeted mRNA-probe heterodimer during digestion, anything not hybridized
with the probe remains unprotected [102]. After digestion, the reaction is precipitated,
gel electrophoresed, and detected by autoradiography [103]. This technique can detect
the presence and abundance of a specific transcript variant, but it is not applicable to
genome-wide applications.

3.3. PCR-Based Techniques

3.3.1. RACE PCR
Rapid Amplification of cDNA Ends (RACE PCR) is a technique that is used to deter-
mine the full-length sequence of an mRNA as long as part of the transcript sequence is
known [104]. RACE PCR is efficient and inexpensive; it requires a cDNA synthesis and a
single PCR reaction to amplify the desired 5′ - or 3′ -end of a certain cDNA of interest. Two
types of RACE PCR are in use: 5′ RACE or 3′ RACE. The first step for both procedures is to
utilize reverse transcription to create a cDNA copy of a region of RNA transcript [105].
The 5′ RACE approaches are slightly more complex than the 3′ RACE since the 5′ -ends
on the mRNAs do not have generic priming sites, as seen in the poly(A) tail [106]. The
addition of an adapter sequence at the 5′ -end will lead to accurate characterizations of
the 5′ UTR and completion of the 5′ RACE. Current methodologies take advantage of the
MMLV reverse transcriptase during first-stand cDNA synthesis [107]. Typically, 2–4 non-
templated cytosine residues are added to the 3′ -end of synthesized cDNAs once the mRNA
reaches the 5′ -end, cap region. Terminal 3–4 G residues base pairs with 2–4 C residues of
the created cDNA if an oligonucleotide with oligo(G) or oligo(rG) sequences is included in
the incubation medium [105]. The new template will elicit a reverse transcriptase template
switch and replicate the sequence of the oligonucleotide [105]. As the MMLV reverse
transcriptase adds C residues to the cDNA, whole cDNA libraries are amplified with
full-length clones. A homopolymer tail is then joined to the cDNA end using terminal
deoxynucleotidyl transferase, which can bind a universal primer during downstream
PCR [108]. Then, PCR is performed using a reverse gene-specific primer and a universal
 (UAP1). These primers amplify cDNA, generating a product that includes the un
5′ sequence.
Genes 2025, 16, 343
In 3′ RACE, an oligo-dT-containing primer complementary to the poly(A) t
11 of 18
cluding an adapter oligo sequence) is used to bind to mRNA transcripts and revers
scribed to synthesize the 3′-end of the cDNA [109]. Then, PCR is performed using
primer (UAP1). These primers amplify cDNA, generating a product that includes the
specific forward primer that binds to a known cDNA sequence and a primer that b
unknown 5′ sequence.
the adapter sequence [110]. This process produces a product that includes the sequ
In 3′ RACE, an oligo-dT-containing primer complementary to the poly(A) tail (includ-
interest at theoligo
ing an adapter 3′-end. Afteris the
sequence) usedinitial
to bind cycle of PCR
to mRNA for both
transcripts 5′ RACE
and reverse and 3′ RACE
transcribed
tional PCR cycles
to synthesize ′ areof
the 3 -end carried
the cDNAout[109].
using nested
Then, PCRprimers to increase
is performed specificity [111].
using a gene-specific
cases, a known section of a specific mRNA transcript is used to create a to
forward primer that binds to a known cDNA sequence and a primer that binds the of th
cDNA
adapter sequence [110]. This process produces a product that includes the sequence of
transcript sequence, which is then amplified during PCR. Thus, this process can de
interest at the 3′ -end. After the initial cycle of PCR for both 5′ RACE and 3′ RACE, addi-
unknown sequences of a single mRNA transcript variant.
tional PCR cycles are carried out using nested primers to increase specificity [111]. In both
cases, a known section of a specific mRNA transcript is used to create a cDNA of the entire
3.3.2. RT-PCR
transcript andwhich
sequence, RT-qPCR
is then amplified during PCR. Thus, this process can detect the
unknown sequences
The first step of
ofathe
single mRNAtranscription
reverse transcript variant.
polymerase chain reaction (RT-PCR
reverse transcription
3.3.2. RT-PCR and RT-qPCR of mRNA to generate cDNA. cDNA is used as a template
downstream PCR
The first step reaction
of the reverseto amplify the
transcription cDNA sequences.
polymerase chain reactionAmplified
(RT-PCR) isPCR
the produ
be visualized
reverse through
transcription gel electrophoresis,
of mRNA to generate cDNA. followed
cDNA isbyusedethidium bromide
as a template for theor SYBR
downstream PCR reaction to amplify the cDNA sequences. Amplified
staining. RT-PCR is a widely used method to determine the presence of specific PCR products can RN
be visualized through gel electrophoresis, followed by ethidium bromide or SYBR Green
ments, leading to discoveries in gene expression. This method can also be applied to
staining. RT-PCR is a widely used method to determine the presence of specific RNA
transcript variants;
segments, leading specific primers
to discoveries can be designed
in gene expression. to bind
This method can to
alsovariant-specific
be applied ex
regions, flanking variants;
to detect transcript alternatively
specificspliced
primersregions to amplify
can be designed thetopresence
to bind of a specific
variant-specific
(Figure
exons or 4). Thisflanking
regions, allowsalternatively
researchers to identify
spliced regions towhich
amplifyvariants areofpresent
the presence a specific in the
and further our understanding of the variants’ functional diversity. in the
variant (Figure 4). This allows researchers to identify which variants are present
sample and further our understanding of the variants’ functional diversity.

4. Detection
Figure 4.
Figure Detection of mRNA transcript
of mRNA variantsvariants
transcript using RT-PCR.
usingART-PCR.
schematicA
illustrates
schematicthe exon
illustrates t
composition of the Runx1 transcript variants. RT-PCR can be performed using primers designed for
composition of the Runx1 transcript variants. RT-PCR can be performed using primers desig
variant-specific exons. Using the forward (Fd) and reverse (Rv) primers, Runx1-201, 202, and 203
variant-specific exons.
can be detected, but Using the
the remaining forward
variants cannot(Fd) and reverse
be detected (Rv)
due to the primers,
failure Runx1-201,
of primer binding. 202,
Moreover,
can RT-PCRbut
be detected, willthe
fail remaining
to differentiate Runx1-201
variants and be
cannot Runx1-203.
detectedSolid
duearrows
to the indicate theprimer b
failure of
binding of primers to target templates, whereas dotted arrows indicate an inability to bind.
Moreover, RT-PCR will fail to differentiate Runx1-201 and Runx1-203. Solid arrows indi
binding of primers
RT-qPCR to target
is a process like templates,
RT-PCR, butwhereas dotted
its primary arrows
purpose indicate an
is quantifying inability
cDNAs. The to bind.
procedure for RT-qPCR fluorescent dyes or probes monitors the accumulation of the PCR
product in real time.
RT-qPCR is aThis process
process easily
like allows the
RT-PCR, butquantifying
its primaryof specific
purpose cDNAs. It can
is quantifying
c
also be applied to detect transcript variants, as it can quantify levels of different transcript
The procedure for RT-qPCR fluorescent dyes or probes monitors the accumulation
PCR product in real time. This process easily allows the quantifying of specific cD
can also be applied to detect transcript variants, as it can quantify levels of differen
script variants. Using specific primers or probes unique to each variant, research
Genes 2025, 16, 343 12 of 18

variants. Using specific primers or probes unique to each variant, researchers can measure
the relative abundance of each variant in a sample. As RT-PCR and RT-qPCR can efficiently
detect selective mRNA transcript variants, these techniques are often used to verify RNA
sequencing results.

3.4. Machine Learning

The recent advancement in the field of variant detection comes in the form of prediction
algorithms in machine learning. Studies have used splice prediction tools, although there is
still inadequate consensus for a tool to be optimal. MaxEntScan (MES) is an older method
that is based on a maximum entropy model to capture the highest score as the best fit for a
potential donor or accept site [112]. Another sliding window algorithm called MES-SWA
can be used to capture the transcript variants. Modular Modeling of Splicing (MMSplice)
is a software built on multiple neural networks and deep learning frameworks trained
to score exon, intron, and other splice sites to predict transcript variants [113]. Super
Quick Information-content Random-forest Learning of Splice variants (SQUIRLS) stands
out as a unique approach to interpreting nucleotide changes at the 3′ and 5′ UTRs as it
generates interpretable features through its engineered decision trees to classify splice
sites [113,114]. SpliceAI is another AI tool that was trained initially based on human
reference genome data and used for the identification of splice junctions [115,116]. It
uses neural networks that work on 10,000 nucleotides of context to predict splice sites. A
collapsed isoform (CI) set representing manually annotated constitutive and alternative
splice sites was included in SpliceAI to improve prediction. The CI-SpliceAI gained more
accuracy in predictions [115,117,118]. ASTK, a recent software package that covers both the
downstream and upstream analysis of alternative splicing, provides enrichment analysis
at the gene and exon levels. This package extracts the specific features of the targeted
sequence, as well as its epigenetic marks that align with splicing events, uncovering that
splice strength is a determinant for A3 and A5 exon inclusion levels [119].
Methods that enhance machine learning algorithms for a better understanding of
splice variant function and alternative splicing events are still improving [120]. The in-
creased ability to locate the motifs of RNA-binding proteins and their splicing effects will
improve the detection of alternative splicing. Including newer elements, such as epigenetic
marks, in these algorithms will also play an essential role in discovering the intricacies of
splice regulation.

4. Advantages and Disadvantages of Detection Methods

The applicability of any detection technique will depend on the efficiency of sequenc-
ing the whole mRNA from the 5′ -end to the 3′ -end of mRNAs. Considering the available
techniques, RNA-Seq is the best for detecting mRNA transcript variants. Illumina, PacBio,
and Nanopore technologies are the most popular RNA-Seq techniques [36]. Other sequenc-
ing methods include scRNA-Seq using 10x Genomics, Takara Bio, or Bio-Rad systems.
Illumina sequencing is the most accurate and cost-effective NGS platform, making it suit-
able for research. However, despite its high accuracy, RNA-Seq on the Illumina platform
suffers from de novo assembly and alignment problems of the short reads. In this regard,
long-read sequences on nanopore or PacBio appear better. The key advantage of nanopore
or PacBio sequencing is its ability to perform long reads up to hundreds of kilobases. These
techniques can sequence RNAs without PCR amplification, reducing PCR biases and errors.
However, it must be noted that the random error rate of base reading in nanopore or
PacBio (5–15%) is remarkably higher than that of Illumina (0.1–0.5%) [121]. The initial
establishment cost of the Illumina or PacBio system is much higher than that of nanopore
technologies. However, the sequencing cost per sample is much higher in nanopore or
Genes 2025, 16, 343 13 of 18

PacBio systems. Newer high-throughput sequencing systems and reagent chemistry have
lowered the cost of nanopore sequencing, which is still higher than Illumina systems [122].
Direct RNA-Seq using nanopore technology remains the best laboratory technique for
detecting mRNA transcript variants. However, these bulk RNA sequences can represent
a tissue or organ but cannot identify the cellular origin of specific transcript variants.
scRNA-Seq can identify the cellular origin of the transcript, but commonly used scRNA-
Seq (e.g., 10x Genomics), which targets the 3′ -end of mRNA only, is not useful for analyzing
transcript variants. However, mRNA sequencing of isolated single cells (Takara bio or
low-input RNA-seq) or full-length RNA-Seq in single cells (using combined nanopore
and 10x Genomics) is applicable for mRNA transcript variant detection. However, these
methods may exhibit a low depth of RNA sequencing.
Another limitation remains with analyzing RNA-Seq data. RNA-Seq performed on the
Illumina platform can be analyzed using many open or commercial software programs. In
contrast, only a limited number of software programs are available to analyze RNA-Seq data
from other platforms. Despite the labor-intensive procedures, hybridization-based methods
are not as efficient as RNA-Seq methods in detecting mRNA sequence variants [123].
However, PCR-based methods have selective advantages and can complement the RNA
sequencing approach [32]. RT-PCR and Sanger sequencing can be used to verify any RNA
sequencing data.

5. Conclusions and Future Perspectives

Only one mRNA transcript expressed from a gene is often focused on in gene expres-
sion studies. However, this simplistic approach is biologically inaccurate; most mammalian
genes express more than one mature mRNA. The major barrier in transcript variant analysis
is the efficacy and cost-effectiveness of the detection methods. Considering the available
laboratory techniques, RNA-Seq is the best for detecting mRNA transcript variants. Com-
pared to short-read-based sequencing using the Illumina platform, long-read sequences on
ONT or PacBio platforms appear better due to de novo assembly and alignment advantages.
However, long-read sequences and direct mRNA sequencing require large quantities of
RNA, which is feasible for bulk sequencing. Bulk sequencing with long reads can identify
mRNA transcript variants efficiently, but it fails to determine the cellular origin of a particu-
lar transcript variant. scRNA-Seq can resolve the issue of cell type identification. However,
scRNA-Seq typically detects the 3′ -end of mRNAs. Researchers have addressed this issue
by combining ONT or PacBio sequencing (long read) with 10x Genomics-based scRNA-Seq.
We anticipate that future studies will be directed toward developing scRNA-Seq techniques
that perform long-read direct mRNA sequencing.

Author Contributions: M.A.K.R. conceptualized, supervised, provided resources, and edited; K.V.,
S.S., Y.S., G.J.P., C.Y.R., and V.D. wrote the original manuscript; all authors read and agreed on the
manuscript’s contents. P.E.F. contributed to reviewing and editing the manuscript. All authors have
read and agreed to the published version of the manuscript.

Funding: The Department of Pathology and Laboratory Medicine at the University of Kansas Medical
Center partially supported K.V., P.E.F., and M.A.K.R. No institutional financing was involved.

Institutional Review Board Statement: This study did not include humans or animals.

Conflicts of Interest: The authors declare no conflicts of interest.

Genes 2025, 16, 343 14 of 18

References
1. Santosh, B.; Varshney, A.; Yadava, P.K. Non-coding RNAs: Biological functions and applications. Cell Biochem. Funct. 2015, 33, 14–22.
[CrossRef]
2. Statello, L.; Guo, C.-J.; Chen, L.-L.; Huarte, M. Gene regulation by long non-coding RNAs and its biological functions. Nat. Rev.
Mol. Cell Biol. 2021, 22, 96–118. [CrossRef]
3. Ma, B.; Wang, S.; Wu, W.; Shan, P.; Chen, Y.; Meng, J.; Xing, L.; Yun, J.; Hao, L.; Wang, X.; et al. Mechanisms of circRNA/lncRNA-
miRNA interactions and applications in disease and drug research. Biomed. Pharmacother. = Biomed. Pharmacother. 2023, 162, 114672.
[CrossRef]
4. Redi, C.A.; Capanna, E. Genome size evolution: Sizing mammalian genomes. Cytogenet. Genome Res. 2012, 137, 97–112. [CrossRef]
[PubMed]
5. Lee, Y.; Rio, D.C. Mechanisms and regulation of alternative pre-mRNA splicing. Annu. Rev. Biochem. 2015, 84, 291–323. [CrossRef]
6. Zhong, W.; Wu, Y.; Zhu, M.; Zhong, H.; Huang, C.; Lin, Y.; Huang, J. Alternative splicing and alternative polyadenylation define
tumor immune microenvironment and pharmacogenomic landscape in clear cell renal carcinoma. Mol. Ther. Nucleic Acids
2022, 27, 927–946. [CrossRef] [PubMed]
7. Marasco, L.E.; Kornblihtt, A.R. The physiology of alternative splicing. Nat. Rev. Mol. Cell Biol. 2023, 24, 242–254. [CrossRef]
[PubMed]
8. Shabalina, S.A.; Spiridonov, N.A. The mammalian transcriptome and the function of non-coding DNA sequences. Genome Biol.
2004, 5, 105. [CrossRef]
9. de Sousa Abreu, R.; Penalva, L.O.; Marcotte, E.M.; Vogel, C. Global signatures of protein and mRNA expression levels. Mol. Biosyst.
2009, 5, 1512–1526. [CrossRef]
10. Alberts, B.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walter, P. From DNA to RNA. In Molecular Biology of the Cell, 4th ed.;
Garland Science: New York, NY, USA, 2002.
11. Vo, K.; Sharma, Y.; Paul, A.; Mohamadi, R.; Mohamadi, A.; Fields, P.E.; Rumi, M.K. Importance of transcript variants in
transcriptome analyses. Cells 2024, 13, 1502. [CrossRef]
12. Sharma, Y.; Vo, K.; Shila, S.; Paul, A.; Dahiya, V.; Fields, P.E.; Rumi, M.A.K. mRNA Transcript Variants Expressed in Mammalian
Cells. Int. J. Mol. Sci. 2025, 26, 1052. [CrossRef]
13. Schwanhäusser, B.; Busse, D.; Li, N.; Dittmar, G.; Schuchhardt, J.; Wolf, J.; Chen, W.; Selbach, M. Global quantification of
mammalian gene expression control. Nature 2011, 473, 337–342. [CrossRef] [PubMed]
14. Kochetov, A.V. Alternative translation start sites and hidden coding potential of eukaryotic mRNAs. Bioessays 2008, 30, 683–691.
[CrossRef]
15. Di Giammartino, D.C.; Nishida, K.; Manley, J.L. Mechanisms and consequences of alternative polyadenylation. Mol. Cell 2011, 43,
853–866. [CrossRef] [PubMed]
16. Mohanan, N.K.; Shaji, F.; Koshre, G.R.; Laishram, R.S. Alternative polyadenylation: An enigma of transcript length variation in
health and disease. Wiley Interdiscip. Rev. RNA 2022, 13, e1692. [CrossRef] [PubMed]
17. Ayoubi, T.A.; Van De Ven, W.J. Regulation of gene expression by alternative promoters. FASEB J 1996, 10, 453–460. [CrossRef]
18. Komeno, Y.; Yan, M.; Matsuura, S.; Lam, K.; Lo, M.-C.; Huang, Y.-J.; Tenen, D.G.; Downing, J.R.; Zhang, D.-E. Runx1 exon 6–related
alternative splicing isoforms differentially regulate hematopoiesis in mice. Blood J. Am. Soc. Hematol. 2014, 123, 3760–3769. [CrossRef]
19. Warren, C.F.A.; Wong-Brown, M.W.; Bowden, N.A. BCL-2 family isoforms in apoptosis and cancer. Cell Death Dis. 2019, 10, 177.
[CrossRef]
20. Keller, M.A.; Huang, C.-y.; Ivessa, A.; Singh, S.; Romanienko, P.J.; Nakamura, M. Bcl-x short-isoform is essential for maintaining
homeostasis of multiple tissues. iScience 2023, 26, 106409. [CrossRef] [PubMed]
21. Sheynkman, G.M.; Tuttle, K.S.; Laval, F.; Tseng, E.; Underwood, J.G.; Yu, L.; Dong, D.; Smith, M.L.; Sebra, R.; Willems, L.; et al.
ORF Capture-Seq as a versatile method for targeted identification of full-length isoforms. Nat. Commun. 2020, 11, 2326. [CrossRef]
22. Kovacs, E.; Tompa, P.; Liliom, K.; Kalmar, L. Dual coding in alternative reading frames correlates with intrinsic protein disorder.
Proc. Natl. Acad. Sci. USA 2010, 107, 5429–5434. [CrossRef]
23. Dhamija, S.; Menon, M.B. Non-coding transcript variants of protein-coding genes—What are they good for? RNA Biol. 2018, 15,
1025–1031. [CrossRef] [PubMed]
24. Potter, S.S. Single-cell RNA sequencing for the study of development, physiology and disease. Nat. Rev. Nephrol. 2018, 14, 479–492.
[CrossRef] [PubMed]
25. Yang, H.D.; Nam, S.W. Pathogenic diversity of RNA variants and RNA variation-associated factors in cancer development.
Exp. Mol. Med. 2020, 52, 582–593. [CrossRef]
26. Ward, A.J.; Cooper, T.A. The pathobiology of splicing. J. Pathol. J. Pathol. Soc. Great Br. Irel. 2010, 220, 152–163. [CrossRef]
[PubMed]
27. Tazi, J.; Bakkour, N.; Stamm, S. Alternative splicing and disease. Biochim. Biophys. Acta (BBA)-Mol. Basis Dis. 2009, 1792, 14–26.
[CrossRef]
Genes 2025, 16, 343 15 of 18

28. Gimeno-Valiente, F.; López-Rodas, G.; Castillo, J.; Franco, L. Alternative splicing, epigenetic modifications and cancer:
A dangerous triangle, or a hopeful one? Cancers 2022, 14, 560. [CrossRef]
29. Kwan, T.; Benovoy, D.; Dias, C.; Gurd, S.; Provencher, C.; Beaulieu, P.; Hudson, T.J.; Sladek, R.; Majewski, J. Genome-wide analysis
of transcript isoform variation in humans. Nat. Genet. 2008, 40, 225–231. [CrossRef]
30. Haraksingh, R.R.; Snyder, M.P. Impacts of variation in the human genome on gene regulation. J. Mol. Biol. 2013, 425, 3970–3977.
[CrossRef]
31. Chen, B.; Scurrah, C.R.; McKinley, E.T.; Simmons, A.J.; Ramirez-Solano, M.A.; Zhu, X.; Markham, N.O.; Heiser, C.N.; Vega,
P.N.; Rolong, A.; et al. Differential pre-malignant programs and microenvironment chart distinct paths to malignancy in human
colorectal polyps. Cell 2021, 184, 6262–6280.e26. [CrossRef]
32. Kukurba, K.R.; Montgomery, S.B. RNA Sequencing and Analysis. Cold Spring Harb. Protoc. 2015, 2015, 951–969. [CrossRef]
33. Parkhomchuk, D.; Borodina, T.; Amstislavskiy, V.; Banaru, M.; Hallen, L.; Krobitsch, S.; Lehrach, H.; Soldatov, A. Transcriptome
analysis by strand-specific sequencing of complementary DNA. Nucleic Acids Res. 2009, 37, e123. [CrossRef] [PubMed]
34. Wei, X.; Wang, X. A computational workflow to identify allele-specific expression and epigenetic modification in maize.
Genom. Proteom. Bioinform. 2013, 11, 247–252. [CrossRef] [PubMed]
35. Xiong, Y.; Soumillon, M.; Wu, J.; Hansen, J.; Hu, B.; Van Hasselt, J.G.; Jayaraman, G.; Lim, R.; Bouhaddou, M.; Ornelas, L.
A comparison of mRNA sequencing with random primed and 3′ -directed libraries. Sci. Rep. 2017, 7, 14626. [CrossRef] [PubMed]
36. Hu, T.; Chitnis, N.; Monos, D.; Dinh, A. Next-generation sequencing technologies: An overview. Human. Immunol. 2021, 82, 801–811.
[CrossRef]
37. Jain, M.; Abu-Shumays, R.; Olsen, H.E.; Akeson, M. Advances in nanopore direct RNA sequencing. Nat. Methods 2022, 19, 1160–1164.
[CrossRef]
38. De Maio, N.; Shaw, L.P.; Hubbard, A.; George, S.; Sanderson, N.D.; Swann, J.; Wick, R.; AbuOun, M.; Stubberfield, E.; Hoosdally,
S.J. Comparison of long-read sequencing technologies in the hybrid assembly of complex bacterial genomes. Microb. Genom.
2019, 5, e000294. [CrossRef]
39. Mao, S.; Su, J.; Wang, L.; Bo, X.; Li, C.; Chen, H. A transcriptome-based single-cell biological age model and resource for
tissue-specific aging measures. Genome Res. 2023, 33, 1381–1394. [CrossRef]
40. Li, X.; Wang, C.-Y. From bulk, single-cell to spatial RNA sequencing. Int. J. Oral Sci. 2021, 13, 36. [CrossRef]
41. Wu, Y.; Zhang, K. Tools for the analysis of high-dimensional single-cell RNA sequencing data. Nat. Rev. Nephrol. 2020, 16, 408–421.
[CrossRef]
42. Chen, G.; Ning, B.; Shi, T. Single-Cell RNA-Seq Technologies and Related Computational Data Analysis. Front. Genet. 2019, 10, 317.
[CrossRef]
43. Ren, X.; Kang, B.; Zhang, Z. Understanding tumor ecosystems by single-cell sequencing: Promises and limitations. Genome Biol.
2018, 19, 211. [CrossRef] [PubMed]
44. Zhong, S.; Joung, J.-G.; Zheng, Y.; Chen, Y.-R.; Liu, B.; Shao, Y.; Xiang, J.Z.; Fei, Z.; Giovannoni, J.J. High-throughput illumina
strand-specific RNA sequencing library preparation. Cold Spring Harb. Protoc. 2011, 2011, 940–949. [CrossRef] [PubMed]
45. Vivancos, A.P.; Güell, M.; Dohm, J.C.; Serrano, L.; Himmelbauer, H. Strand-specific deep sequencing of the transcriptome. Genome
Res. 2010, 20, 989–999. [CrossRef] [PubMed]
46. Stark, R.; Grzelak, M.; Hadfield, J. RNA sequencing: The teenage years. Nat. Rev. Genet. 2019, 20, 631–656. [CrossRef]
47. Dorney, R.; Dhungel, B.P.; Rasko, J.E.J.; Hebbard, L.; Schmitz, U. Recent advances in cancer fusion transcript detection. Brief. Bioinform.
2022, 24, bbac519. [CrossRef]
48. Wongsurawat, T.; Jenjaroenpun, P.; Nookaew, I. Direct Sequencing of RNA and RNA Modification Identification Using Nanopore.
Methods Mol. Biol. 2022, 2477, 71–77. [CrossRef]
49. Deng, E.; Shen, Q.; Zhang, J.; Fang, Y.; Chang, L.; Luo, G.; Fan, X. Systematic evaluation of single-cell RNA-seq analyses
performance based on long-read sequencing platforms. J. Adv. Res. 2024, 210–218. [CrossRef]
50. Viscardi, M.J.; Arribere, J.A. Poly (a) selection introduces bias and undue noise in direct RNA-sequencing. BMC Genom. 2022, 23, 530.
[CrossRef]
51. Uapinyoying, P.; Goecks, J.; Knoblach, S.M.; Panchapakesan, K.; Bonnemann, C.G.; Partridge, T.A.; Jaiswal, J.K.; Hoffman, E.P.
A long-read RNA-seq approach to identify novel transcripts of very large genes. Genome Res. 2020, 30, 885–897. [CrossRef]
52. Gehrig, J.L.; Portik, D.M.; Driscoll, M.D.; Jackson, E.; Chakraborty, S.; Gratalo, D.; Ashby, M.; Valladares, R. Finding the right
fit: Evaluation of short-read and long-read sequencing approaches to maximize the utility of clinical microbiome data. Microb.
Genom. 2022, 8, 000794. [CrossRef] [PubMed]
53. Gong, B.; Li, D.; Łabaj, P.P.; Pan, B.; Novoradovskaya, N.; Thierry-Mieg, D.; Thierry-Mieg, J.; Chen, G.; Bergstrom Lucas, A.;
LoCoco, J.S. Targeted DNA-seq and RNA-seq of Reference Samples with Short-read and Long-read Sequencing. Sci. Data
2024, 11, 892. [CrossRef] [PubMed]
54. McCarty, D.M.; Young Jr, S.M.; Samulski, R.J. Integration of adeno-associated virus (AAV) and recombinant AAV vectors.
Annu. Rev. Genet. 2004, 38, 819–845. [CrossRef]
Genes 2025, 16, 343 16 of 18

55. Despic, V.; Jaffrey, S.R. mRNA ageing shapes the Cap2 methylome in mammalian mRNA. Nature 2023, 614, 358–366. [CrossRef]
[PubMed]
56. Leger, A.; Amaral, P.P.; Pandolfini, L.; Capitanchik, C.; Capraro, F.; Miano, V.; Migliori, V.; Toolan-Kerr, P.; Sideri, T.; Enright, A.J.;
et al. RNA modifications detection by comparative Nanopore direct RNA sequencing. Nat. Commun. 2021, 12, 7198. [CrossRef]
57. Ruiz, A.; Bok, D. Direct RT-PCR amplification of mRNA supported on membranes. Biotechniques 1993, 15, 882–887.
58. Aird, D.; Ross, M.G.; Chen, W.S.; Danielsson, M.; Fennell, T.; Russ, C.; Jaffe, D.B.; Nusbaum, C.; Gnirke, A. Analyzing and
minimizing PCR amplification bias in Illumina sequencing libraries. Genome Biol. 2011, 12, R18. [CrossRef]
59. Verwilt, J.; Mestdagh, P.; Vandesompele, J. Artifacts and biases of the reverse transcription reaction in RNA sequencing. RNA
2023, 29, 889–897. [CrossRef]
60. Parekh, S.; Ziegenhain, C.; Vieth, B.; Enard, W.; Hellmann, I. The impact of amplification on differential expression analyses by
RNA-seq. Sci. Rep. 2016, 6, 25533. [CrossRef]
61. Kebschull, J.M.; Zador, A.M. Sources of PCR-induced distortions in high-throughput sequencing data sets. Nucleic Acids Res.
2015, 43, e143. [CrossRef]
62. Kivioja, T.; Vähärautio, A.; Karlsson, K.; Bonke, M.; Enge, M.; Linnarsson, S.; Taipale, J. Counting absolute numbers of molecules
using unique molecular identifiers. Nat. Methods 2011, 9, 72–74. [CrossRef] [PubMed]
63. Islam, S.; Zeisel, A.; Joost, S.; La Manno, G.; Zajac, P.; Kasper, M.; Lönnerberg, P.; Linnarsson, S. Quantitative single-cell RNA-seq
with unique molecular identifiers. Nat. Methods 2014, 11, 163–166. [CrossRef]
64. Ikeda, H.; Miyao, S.; Yamada, N.; Sugimoto, S.; Kimura, F.; Kurimoto, K. Protocol for high-quality single-cell RNA-seq from tissue
sections with DRaqL. STAR Protoc. 2024, 5, 103050. [CrossRef]
65. Slovin, S.; Carissimo, A.; Panariello, F.; Grimaldi, A.; Bouché, V.; Gambardella, G.; Cacchiarelli, D. Single-cell RNA sequencing
analysis: A step-by-step overview. RNA Bioinform. 2021, 2284, 343–365.
66. Gao, C.; Zhang, M.; Chen, L. The comparison of two single-cell sequencing platforms: BD rhapsody and 10x genomics chromium.
Curr. Genom. 2020, 21, 602–609. [CrossRef] [PubMed]
67. Haque, A.; Engel, J.; Teichmann, S.A.; Lönnberg, T. A practical guide to single-cell RNA-sequencing for biomedical research and
clinical applications. Genome Med. 2017, 9, 75. [CrossRef]
68. Nguyen, A.; Khoo, W.H.; Moran, I.; Croucher, P.I.; Phan, T.G. Single cell RNA sequencing of rare immune cell populations. Front.
Immunol. 2018, 9, 1553. [CrossRef]
69. Picelli, S.; Faridani, O.R.; Björklund, Å.K.; Winberg, G.; Sagasser, S.; Sandberg, R. Full-length RNA-seq from single cells using
Smart-seq2. Nat. Protoc. 2014, 9, 171–181. [CrossRef]
70. Westoby, J.; Artemov, P.; Hemberg, M.; Ferguson-Smith, A. Obstacles to detecting isoforms using full-length scRNA-seq data.
Genome Biol. 2020, 21, 74. [CrossRef]
71. Yu, X.; Abbas-Aghababazadeh, F.; Chen, Y.A.; Fridley, B.L. Statistical and Bioinformatics Analysis of Data from Bulk and
Single-Cell RNA Sequencing Experiments. Methods Mol. Biol. 2021, 2194, 143–175. [CrossRef]
72. Levin, J.Z.; Yassour, M.; Adiconis, X.; Nusbaum, C.; Thompson, D.A.; Friedman, N.; Gnirke, A.; Regev, A. Comprehensive
comparative analysis of strand-specific RNA sequencing methods. Nat. Methods 2010, 7, 709–715. [CrossRef]
73. Sharma, P.; Sharma, B.S.; Verma, R.J. A Guide to RNAseq Data Analysis Using Bioinformatics Approaches. In Advances in
Bioinformatics; Singh, V., Kumar, A., Eds.; Springer: Singapore, 2021; pp. 243–260.
74. Engström, P.G.; Steijger, T.; Sipos, B.; Grant, G.R.; Kahles, A.; Alioto, T.; Behr, J.; Bertone, P.; Bohnert, R.; Campagna, D.; et al.
Systematic evaluation of spliced alignment programs for RNA-seq data. Nat. Methods 2013, 10, 1185–1191. [CrossRef] [PubMed]
75. Soneson, C.; Love, M.I.; Robinson, M.D. Differential analyses for RNA-seq: Transcript-level estimates improve gene-level
inferences. F1000Research 2015, 4, 1521. [CrossRef]
76. Mortazavi, A.; Williams, B.A.; McCue, K.; Schaeffer, L.; Wold, B. Mapping and quantifying mammalian transcriptomes by
RNA-Seq. Nat. Methods 2008, 5, 621–628. [CrossRef] [PubMed]
77. Emilsson, V.; Thorleifsson, G.; Zhang, B.; Leonardson, A.S.; Zink, F.; Zhu, J.; Carlson, S.; Helgason, A.; Walters, G.B.; Gunnarsdottir,
S.; et al. Genetics of gene expression and its effect on disease. Nature 2008, 452, 423–428. [CrossRef] [PubMed]
78. Trapnell, C.; Williams, B.A.; Pertea, G.; Mortazavi, A.; Kwan, G.; van Baren, M.J.; Salzberg, S.L.; Wold, B.J.; Pachter, L. Transcript
assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation.
Nat. Biotechnol. 2010, 28, 511–515. [CrossRef]
79. Piskol, R.; Ramaswami, G.; Li, J.B. Reliable identification of genomic variants from RNA-seq data. Am. J. Hum. Genet. 2013, 93, 641–651.
[CrossRef]
80. Deshpande, D.; Chhugani, K.; Chang, Y.; Karlsberg, A.; Loeffler, C.; Zhang, J.; Muszyńska, A.; Munteanu, V.; Yang, H.; Rotman, J.
RNA-seq data science: From raw data to effective interpretation. Front. Genet. 2023, 14, 997383. [CrossRef]
81. Su, Y.; Yu, Z.; Jin, S.; Ai, Z.; Yuan, R.; Chen, X.; Xue, Z.; Guo, Y.; Chen, D.; Liang, H.; et al. Comprehensive assessment of mRNA
isoform detection methods for long-read sequencing data. Nat. Commun. 2024, 15, 3972. [CrossRef]
Genes 2025, 16, 343 17 of 18

82. Lebrigand, K.; Bergenstråhle, J.; Thrane, K.; Mollbrink, A.; Meletis, K.; Barbry, P.; Waldmann, R.; Lundeberg, J. The spatial
landscape of gene expression isoforms in tissue sections. Nucleic Acids Res. 2023, 51, e47. [CrossRef]
83. Method of the Year 2020: Spatially resolved transcriptomics. Nat. Methods 2021, 18, 1. [CrossRef] [PubMed]
84. Williams, C.G.; Lee, H.J.; Asatsuma, T.; Vento-Tormo, R.; Haque, A. An introduction to spatial transcriptomics for biomedical
research. Genome Med. 2022, 14, 68. [CrossRef] [PubMed]
85. Chen, T.Y.; You, L.; Hardillo, J.A.U.; Chien, M.P. Spatial Transcriptomic Technologies. Cells 2023, 12, 2042. [CrossRef]
86. Ke, R.; Mignardi, M.; Pacureanu, A.; Svedlund, J.; Botling, J.; Wählby, C.; Nilsson, M. In situ sequencing for RNA analysis in
preserved tissue and cells. Nat. Methods 2013, 10, 857–860. [CrossRef] [PubMed]
87. Gyllborg, D.; Langseth, C.M.; Qian, X.; Choi, E.; Salas, S.M.; Hilscher, M.M.; Lein, E.S.; Nilsson, M. Hybridization-based in situ
sequencing (HybISS) for spatially resolved transcriptomics in human and mouse brain tissue. Nucleic Acids Res. 2020, 48, e112.
[CrossRef]
88. Hilscher, M.M.; Gyllborg, D.; Yokota, C.; Nilsson, M. In Situ Sequencing: A High-Throughput, Multi-Targeted Gene Expression
Profiling Technique for Cell Typing in Tissue Sections. Methods Mol. Biol. 2020, 2148, 313–329. [CrossRef]
89. Sountoulidis, A.; Liontos, A.; Nguyen, H.P.; Firsova, A.B.; Fysikopoulos, A.; Qian, X.; Seeger, W.; Sundström, E.; Nilsson, M.;
Samakovlis, C. SCRINSHOT enables spatial mapping of cell states in tissue sections with single-cell resolution. PLoS Biol. 2020,
18, e3000675. [CrossRef]
90. Lee, J.; Yoo, M.; Choi, J. Recent advances in spatially resolved transcriptomics: Challenges and opportunities. BMB Rep. 2022, 55, 113.
[CrossRef]
91. Yan, K.; Liu, Q.Z.; Huang, R.R.; Jiang, Y.H.; Bian, Z.H.; Li, S.J.; Li, L.; Shen, F.; Tsuneyama, K.; Zhang, Q.L.; et al. Spatial
transcriptomics reveals prognosis-associated cellular heterogeneity in the papillary thyroid carcinoma microenvironment.
Clin. Transl. Med. 2024, 14, e1594. [CrossRef]
92. Zhang, L.; Chen, D.; Song, D.; Liu, X.; Zhang, Y.; Xu, X.; Wang, X. Clinical and translational values of spatial transcriptomics.
Signal Transduct. Target. Ther. 2022, 7, 111. [CrossRef]
93. Niyakan, S.; Sheng, J.; Cao, Y.; Zhang, X.; Xu, Z.; Wu, L.; Wong, S.T.; Qian, X. MUSTANG: Multi-sample spatial transcriptomics
data analysis with cross-sample transcriptional similarity guidance. Patterns 2024, 5, 100986. [CrossRef] [PubMed]
94. Murphy, D. Gene expression studies using microarrays: Principles, problems, and prospects. Adv. Physiol. Educ. 2002, 26, 256–270.
[CrossRef]
95. Lockhart, D.J.; Winzeler, E.A. Genomics, gene expression and DNA arrays. Nature 2000, 405, 827–836. [CrossRef]
96. Yang, T.; Zhang, M.; Zhang, N. Modified Northern blot protocol for easy detection of mRNAs in total RNA using radiolabeled
probes. BMC Genom. 2022, 23, 66. [CrossRef]
97. Rosen, K.M.; Lamperti, E.D.; Villa-Komaroff, L. Optimizing the northern blot procedure. Biotechniques 1990, 8, 398–403.
98. Ouyang, T.; Liu, Z.; Han, Z.; Ge, Q. MicroRNA Detection Specificity: Recent Advances and Future Perspective. Anal. Chem.
2019, 91, 3179–3186. [CrossRef]
99. Carey, M.F.; Peterson, C.L.; Smale, S.T. The RNase protection assay. Cold Spring Harb. Protoc. 2013, 2013, pdb.prot071910.
[CrossRef] [PubMed]
100. Ma, Y.J.; Dissen, G.A.; Rage, F.; Ojeda, S.R. RNase Protection Assay. Methods 1996, 10, 273–278. [CrossRef] [PubMed]
101. Mülhardt, C.; Beese, E.W. Sequences. In Molecular Biology and Genomics; Mülhardt, C., Beese, E.W., Eds.; Academic Press:
Burlington, MA, USA, 2007; pp. 169–221.
102. Rottman, J.B. The Ribonuclease Protection Assay: A Powerful Tool for the Veterinary Pathologist. Vet. Pathol. 2002, 39, 2–9.
[CrossRef]
103. Qu, Y.; Boutjdir, M. RNase protection assay for quantifying gene expression levels. Methods Mol. Biol. 2007, 366, 145–158.
[CrossRef]
104. Frohman, M.A.; Dush, M.K.; Martin, G.R. Rapid production of full-length cDNAs from rare transcripts: Amplification using a
single gene-specific oligonucleotide primer. Proc. Natl. Acad. Sci. USA 1988, 85, 8998–9002. [CrossRef] [PubMed]
105. Schramm, G.; Bruchhaus, I.; Roeder, T. A simple and reliable 5′ -RACE approach. Nucleic Acids Res. 2000, 28, E96. [CrossRef]
[PubMed]
106. Adamopoulos, P.G.; Tsiakanikas, P.; Stolidi, I.; Scorilas, A. A versatile 5′ RACE-Seq methodology for the accurate identification of
the 5′ termini of mRNAs. BMC Genom. 2022, 23, 163. [CrossRef]
107. Bashiardes, S.; Lovett, M. cDNA detection and analysis. Curr. Opin. Chem. Biol. 2001, 5, 15–20. [CrossRef]
108. Lazinski, D.W.; Camilli, A. Homopolymer tail-mediated ligation PCR: A streamlined and highly efficient method for DNA
cloning and library construction. Biotechniques 2013, 54, 25–34. [CrossRef]
109. Frohman, M.A. On beyond classic RACE (rapid amplification of cDNA ends). PCR Methods Appl. 1994, 4, S40–S58. [CrossRef]
110. Ozawa, T.; Kondo, M.; Isobe, M. 3′ rapid amplification of cDNA ends (RACE) walking for rapid structural analysis of large
transcripts. J. Human. Genet. 2004, 49, 102–105. [CrossRef]
Genes 2025, 16, 343 18 of 18

111. Jain, R.; Gomer, R.H.; Murtagh, J.J., Jr. Increasing specificity from the PCR-RACE technique. Biotechniques 1992, 12, 58–59.
[PubMed]
112. Shamsani, J.; Kazakoff, S.H.; Armean, I.M.; McLaren, W.; Parsons, M.T.; Thompson, B.A.; O’Mara, T.A.; Hunt, S.E.; Waddell,
N.; Spurdle, A.B. A plugin for the Ensembl Variant Effect Predictor that uses MaxEntScan to predict variant spliceogenicity.
Bioinformatics 2019, 35, 2315–2317. [CrossRef]
113. Cheng, J.; Nguyen, T.Y.D.; Cygan, K.J.; Çelik, M.H.; Fairbrother, W.G.; Avsec, Ž.; Gagneur, J. MMSplice: Modular modeling
improves the predictions of genetic variant effects on splicing. Genome Biol. 2019, 20, 48. [CrossRef]
114. Barbosa, P.; Savisaar, R.; Carmo-Fonseca, M.; Fonseca, A. Computational prediction of human deep intronic variation. Gigascience
2022, 12, giad085. [CrossRef] [PubMed]
115. Strauch, Y.; Lord, J.; Niranjan, M.; Baralle, D. CI-SpliceAI—Improving machine learning predictions of disease causing splicing
variants using curated alternative splice sites. PLoS ONE 2022, 17, e0269159. [CrossRef] [PubMed]
116. Jónsson, B.A.; Halldórsson, G.H.; Árdal, S.; Rögnvaldsson, S.; Einarsson, E.; Sulem, P.; Guðbjartsson, D.F.; Melsted, P.; Stefánsson,
K.; Úlfarsson, M.Ö. Transformers significantly improve splice site prediction. Commun. Biol. 2024, 7, 1616. [CrossRef]
117. Joglekar, A.P. A Cell-Type Centric View of Alternative Splicing in the Mammalian Brain. Ph.D. Dissertation, Weill Medical College
of Cornell University, New York, NY, USA, 2022.
118. Strauch, Y.L. Improving Diagnosis of Genetic Disease Through Computational Investigation of Splicing. Ph.D. Dissertation,
University of Southampton, Southampton, UK, 2023.
119. Huang, S.; He, J.; Yu, L.; Guo, J.; Jiang, S.; Sun, Z.; Cheng, L.; Chen, X.; Ji, X.; Zhang, Y. ASTK: A Machine Learning-Based
Integrative Software for Alternative Splicing Analysis. Adv. Intell. Syst. 2024, 6, 2300594. [CrossRef]
120. Chen, K.; Zhou, Y.; Ding, M.; Wang, Y.; Ren, Z.; Yang, Y. Self-supervised learning on millions of primary RNA sequences from
72 vertebrates improves sequence-based RNA splicing prediction. Brief. Bioinform. 2024, 25, bbae163. [CrossRef]
121. O’Donnell, C.R.; Wang, H.; Dunbar, W.B. Error analysis of idealized nanopore sequencing. Electrophoresis 2013, 34, 2137–2144.
[CrossRef]
122. Ambardar, S.; Gupta, R.; Trakroo, D.; Lal, R.; Vakhlu, J. High Throughput Sequencing: An Overview of Sequencing Chemistry.
Indian. J. Microbiol. 2016, 56, 394–404. [CrossRef]
123. Wang, Z.; Gerstein, M.; Snyder, M. RNA-Seq: A revolutionary tool for transcriptomics. Nat. Rev. Genet. 2009, 10, 57–63. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.