RU2408011C2 - Method of increasing accuracy of determining sequence of amino acid residues of biopolymer based on mass-spectrometric analysis data, computer system - Google Patents

Abstract

FIELD: chemistry. ^ SUBSTANCE: algorithm of comparing mass spectra with a genome database is applied repeatedly after updating the database with new records or after deleting records from the database or after replacing the database with a database composed of new records. Additional records are generated by making changes in the sequence of identified biopolymers which correspond to replacements, deletions, insertions and modifications of one or more amino acid residues. Present invention also relates to a computer system whose operation is based on the disclosed method. ^ EFFECT: high accuracy of identifying sequence of amino acid residues of a biopolymer. ^ 5 cl, 1 dwg, 1 ex

Description

FIELD OF THE INVENTION

The present invention relates to methods of computer processing of mass spectrometric data, aimed at identifying the primary structure of biopolymers, including proteins and peptides.

BACKGROUND OF THE INVENTION

Computer methods for processing mass spectrometric data, aimed at identifying the primary structure of biopolymers, are currently the main way to conduct research in the field of proteomics.

In the context of the present invention, a biopolymer is considered to be a sequence of amino acid residues encoded in the genome containing at least one peptide bond and capable of containing chemical modifications of the residues, including non-protein components, such as lipids, hydrocarbons, other organic and inorganic elements for example metals. The sequence of amino acid residues is characterized by variability due to the following molecular biological processes: alternative splicing, insertions, deletions and substitutions of single amino acid residues. The last three categories of microvariability of the structure of protein biopolymers are denoted by the abbreviation SAP (Single Aminoacid Polymorphism). The set of individual characteristics of the body's proteins forms its proteotype. To determine the proteotype (proteotyping), a way is needed to identify microheterogeneous differences in the primary structures of proteins.

The identification of the primary structure of biopolymers is based on mass spectrometric data. The term "mass spectrometric data" means information about the mass or mass-charge characteristics of complete proteins, peptide fragments of their hydrolysis, or fragments of induced decay of biopolymer ions. During the preparation of biopolymers for mass spectrometric identification, their primary structure may undergo specific amino acid residues or non-specific modifications, i.e., modifications that are independent of the type of residue in the primary structure of the biopolymer.

Mass spectrometric data processing is performed using bioinformation algorithms. Most of them, for example, the Mowse algorithm [1], are based on a comparison of experimentally obtained mass spectrometric data with calculated estimates based on genomic databases (GDB). “Genomic databases” are a collection of information resources containing information records about the sequences of amino acid residues in proteins obtained by decoding genomic information and (or) deciphering the expressed parts of the genome. The entry in the GDB includes a unique identifier of the protein and the corresponding sequence of amino acid residues in letter coding. When comparing mass spectrometric data with the genomic database, the identification algorithm calculates an estimate of statistical reliability, which makes it possible to judge the probability of correct identification of the protein, taking into account the specified mass spectrometric data and a specific genomic database. A protein is considered identified if the assessment of statistical significance exceeds an arbitrarily set threshold value.

When mass spectrometric identification of a biopolymer occurs, situations arise when some of the mass spectrometric data do not coincide with the GDB, since the latter do not contain information about alternative splicing (AS) and SAP. At the same time, adding additional information about all possible AS and SAP options to the GDB leads to a significant decrease in the level of statistical reliability of identification due to the exponential increase in the combinatorial space of the variants of the primary structures of biopolymers coinciding with the mass spectrometric data [2].

The publication [3] describes a method for increasing the accuracy of determining the amino acid sequence of peptides — protein proteolysis products — according to mass spectrometric analysis based on the use of extended HBD. At the preliminary stage, GDB is expanded by including in it amino acid sequences of proteins containing SAP and post-translational modifications (PTMs) annotated in various sources. At the same time, information on SAP and PTM is searched for for all proteins contained in the source database.

Disclosure of the present invention

The proposed solution to this problem in accordance with the present invention is to re-apply mass spectrometric identification algorithms after entering new records in the GDB, or to create GDB from new records that reflect information about AS and SAP, taking into account the results of protein identification by mass spectrometric data. Thus, the present invention relates to a method for improving the accuracy of determining the sequence of amino acid residues according to mass spectrometric analysis, which involves the use of at least one biopolymer identification algorithm based on a comparison of mass spectrometric data with a genomic database, the algorithm being applied sequentially at least twice.

In accordance with one embodiment, the present invention provides for the primary identification of proteins by the AI algorithm, adding primary structure variants containing AS and SAP products for the identified proteins to the GDB only, and then re-identifying on the enriched database or the same algorithm AI, or another AI algorithm.

In accordance with another embodiment, the present invention provides for the primary identification of proteins by the AI algorithm, the creation of GDB containing the primary structures of AS and SAP products of only previously identified proteins, and then re-identification on the enriched database or the same AI algorithm , or another AI algorithm.

A distinctive advantage of the present invention from similar methods involving the use of a combination of bioinformation algorithms to increase the level of statistical reliability of identification is that identification algorithms are applied sequentially, while the previous algorithm (AI) is coupled with the subsequent one (AI ′) by making changes to the GDB. To implement the proposed method, it is sufficient to use only one mass spectrometric identification algorithm, and not at least two, as, for example, stated in the patent publication [4].

Another distinctive advantage of the present invention from publication [3] is that before each repeated application of the algorithm, changes are made to the GDB, taking into account the results of the previous (their) application (s) of the algorithm (AI). This allows you to significantly increase the efficiency of the search (due to the fact that each subsequent identification is more precise in relation to the previous one (s)) and its reliability (due to a sharp decrease in the probability of obtaining false positive results).

The present invention also relates to a computer system, the operation of which is based on the method disclosed above. The MSD mass spectrometric data are input to the system. These data are used to identify biopolymers by the GDB genomic database using the AI algorithm. Identification results (RI) are a list of protein identifiers for which the assessment of identification reliability exceeds a threshold value set by the user. Primary structure variants are generated for proteins in the composition of radiation sources based on the information contained in external sources of information of the research institute of information about known or suspected AS products and variants of SAP. Specialized databases of genetic polymorphism (e.g., NarMar), databases containing information about known modifications of the protein structure (e.g., UniProt), and also personal data on the results of genotyping (e.g., 23andme.com) can be used as a VII. After that, the AI identification algorithm is used to identify proteins using the GDB database based on the initial MSD mass spectrometric data. The results of the AI ′ algorithm, designated as RI ′, are compared with previous RIs, and it is established which variants of changes in the primary protein structure have been identified.

Brief Description of the Drawing

The drawing shows a diagram of a computing system according to the present invention. In this scheme, the following notation is used:

MSD - initial mass spectrometric data received at the input of the system;

GDB - source genomic database;

AI and AI ′ are mass spectrometric identification algorithms, and it is assumed that AI is identical to AI ';

RI - primary identification results, which are a list of protein identifiers;

RI ′ - re-identification results containing additional protein variants;

MGBD - modification of the genomic database;

GBD ′ is a modified genomic database, which includes variants of proteins contained in external sources of information (VII).

Example 1. Identification of a polymorphic variant of Trypsin-1 protein [Precursor] (Uniprot P07477) by the method according to the present invention

Mass spectrometric data from a study of a sample of human stem cells were downloaded from the Pride system (http://www.ebi.ac.uk/pride/). The initial mass spectrometric identification of the proteins of the loaded mass spectra was performed by the Mascot program using the NCBI-nr database. For one of the identified proteins from the Uniprot database, 13 polymorphic variants were obtained. A new database was created by adding to the NCBI-nr database a list of polymorphic variants of Trypsin-1 protein. Repeated mass spectrometric identification of proteins was carried out by the Mascot program using a new database. As a result of secondary identification, a polymorphic version of the Trypsin-1 [Precursor] protein was identified, which differs from the wild type by the replacement of cysteine at position 139 with phenylalanine. A peptide was identified in the ion fragmentation spectrum

K. (139) FLISGWGNTASSGADYPDELQCLDAPVLSQAK (170) .C containing said single amino acid substitution.

Information sources

[one]. Pappin D.J., Hojrup P., Bleasby A.J., Rapid identification of proteins by peptide-mass fingerprinting, Curr Biol 1993, 3 (6), 327-332.

[2]. Kim S., Gupta N., Pevzner P.A., Spectral probabilities and generating functions of tandem mass spectra: a strike against decoy databases. J Proteome Res 2008, 7, 3354-3363.

[3]. Alves G., Ogurtsov A., Yu Y., RAId_DbS: mass-spectrometry based peptide identification web server with knowledge integration. BMC Genomics 2008, 9, 505.

[four]. Method and system for elucidating the primary structure of biopolymers; Bluggel M., Chamrad D., PROTAGEN AG, Dortmund (DE); United States Patent Application Publication US 2006/0188887 Al, Pub. Date: 08.24.2006.

Claims (5)

1. A method of improving the accuracy of determining the sequence of amino acid residues according to mass spectrometric analysis, which involves the use of at least one biopolymer identification algorithm based on a comparison of mass spectrometric data with a genomic database, the algorithm being sequentially applied at least twice, characterized in that before each repeated application of the algorithm, changes are made to the genomic database, taking into account the results of the previous (them) when change (s). 2. The method according to claim 1, in which, before re-applying the algorithm, additional entries are made to the genomic database. 3. The method according to claim 2, in which additional entries are generated by making changes in the sequence of the identified biopolymers corresponding to substitutions, deletions, inserts and modifications of one or more amino acid residues. 4. The method according to claim 1, in which, before re-applying the algorithm, the genomic database is replaced with a database consisting of records corresponding to previously identified biopolymers, as well as records created by making changes in the sequence of identified biopolymers corresponding to replacements, deletions, inserts and modifications of one or more amino acid residues. 5. A computing system, the operation of which is based on the method according to any one of claims 1 to 4.

Images