A protein mass spectrometry identification method, system, device and storage medium

Abstract

The present invention belongs to the field of bioinformatics, and specifically relates to a protein mass spectrometry identification method, system, device and storage medium, wherein the first sequence composed of theoretical peptide ions and the second sequence composed of mass spectrum peak ions are first screened by the ion matching degree of the first sequence and the second sequence to obtain a third sequence, and then the peptide segments generated by the second sequence are matched with the first sequence, and the peptide segments are screened according to the length of the matched peptide segments to obtain a fourth sequence, and the third sequence and the fourth sequence constitute a candidate sequence, and then the second sequence is roughly aligned and precisely aligned with the candidate sequence to generate a directed acyclic graph, and a candidate protein is obtained according to the directed acyclic graph, and the candidate protein is screened by a scoring function to obtain an identification result. The present application has the effect of improving the recognition accuracy of protein mass spectrometry.

Description

Technical Field

[0001] The present invention belongs to the field of bioinformatics, and in particular relates to a protein mass spectrometry identification method, system, equipment and storage medium. Background Art

[0002] The molecular structure and state changes of proteins (i.e. protein variants) are closely related to the pathological changes of diseases. Protein variants help identify disease diagnostic markers and therapeutic targets, thereby improving the diagnosis and treatment of diseases. It has important practical significance for the development of medicine and human health. Therefore, how to accurately identify protein variants has become a research focus in the field of proteomics.

[0003] In related technologies, with the rapid development of protein separation technology and biomacromolecule mass spectrometry technology, biologists are able to separate single proteins for top-down mass spectrometry analysis, thus being able to study the complete molecular structure and state information of a single protein. Although good results have been achieved, there are limitations such as the difficulty in accurately matching mass spectra and protein sequences, the matching process taking a lot of time, and the difficulty in correctly locating specific amino acid modification sites. To this end, researchers began to study related algorithms with computer thinking to solve such problems, and coupled with high-resolution mass spectrometry technology, it has become possible to identify protein variants using computer algorithms. The current protein variant identification methods based on top-down mass spectrometry technology can be divided into the extended protein variant database method and the blind PSA search method.

[0004] The extended protein variant database method is essentially to automatically enumerate all possible protein variants based on the annotation information in the database. Since the protein variant database contains possible protein variant sequences, the protein variant database-based method matches the mass spectrum with the protein variant sequence in the database to quickly identify protein variants. Extended protein variant database methods include: ProSight, MascotTD, BUPID-top-down, ProteinGoggle, and MetaMorpheus. The ProSight team has developed a series of protein variant characterization tools based on the protein variant database, such as ProSightPTM, ProSightPC, ProSightPD, and ProSightLite. The team adopted an offline "shotgun annotation" database strategy to develop a protein variant database containing sequence mutations, alternative splicing, and PTMs. ProSightPTM identifies complete proteins by searching this protein variant database. ProSightPTM is extended to a commercial software system called ProSightPC, which allows users to search tandem mass spectrometry data in the PTM database. The MascotTD method removes uncommon protein variant sequences from the protein variant database to keep the database size within a certain range, and then uses the removed protein variant database to identify complete protein variants. The BUPID-top-down method can analyze mass spectra through various fragmentation methods, as well as identify internal fragments, side chain losses, neutral losses, and PTMs. For unknown proteins, BUPID-top-down searches the database for the best matching protein sequence and speeds up the calculation through a heuristic model. When customizing the database, the ProteinGoggle method includes user-selected PTMs and amino acid variations in the protein variant database for protein variant identification. MetaMorpheus is an open source proteomics database search software system that adds the function of global PTM discovery to improve the identification performance of protein variants. As the protein variant database expands, the identification performance of this type of method improves, but the required storage space and search time also increase. Due to the large number of theoretical protein variants, the extended library method is difficult to identify PSAs such as uncommon and variable PTMs due to storage space and search time constraints.

[0005] Regarding the above-mentioned related technologies, although the blind PSA search method is superior to the extended protein variant database method in terms of performance, this method is more time-consuming, and the performance of the identification method depends largely on the sensitivity and efficiency of the filtering algorithm combined with it. In addition, during the identification process, the current identification method based on the graph model uses the modification information input by the user as the basis for adding edges in order to reduce the complexity of the graph model, which makes it difficult to find unexpected modification sites and amino acid structure variations. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a protein mass spectrometry identification method, system, device and storage medium to improve the accuracy of protein mass spectrometry identification and shorten the identification time.

[0007] A protein mass spectrometry identification method, comprising:

[0008] According to the protein reference sequence, construct the first sequence composed of theoretical peptide ions, wherein the number of the first sequence is multiple;

[0009] Obtaining a mass spectrum file corresponding to the protein to be identified, and constructing a second sequence consisting of mass spectrum peak ions for each mass spectrum in the mass spectrum file;

[0010] Comparing the masses of ions in the first sequence and the second sequence to obtain the number of ions in the first sequence whose ion masses are within an error range, calculating the ion matching degree between the first sequence and the second sequence according to the number of ions and a matching degree formula, and selecting the first n sequences with the highest ion matching degrees as the third sequence;

[0011] According to the difference between the masses of the mass spectrum peak ions in the second sequence, a corresponding peptide tag is obtained, the peptide tag is matched with a plurality of the first sequences to obtain the longest peptide tag length that each of the first sequences can match, the peptide tag lengths are sorted according to the length, the first sequences corresponding to the first m peptide tags with the longest lengths are selected as the fourth sequence, and the third sequence and the fourth sequence are combined as a candidate sequence;

[0012] Based on a continuous wavelet dynamic time warping algorithm, the second sequence is roughly aligned with the candidate sequence to obtain a mapping relationship between ions of the second sequence and ions of the first sequence, and a roughly aligned sequence is obtained according to the mapping relationship, the second sequence and the first sequence, wherein the mapping relationship is that the ion mass of one or more ions in the second sequence is similar to the ion mass of a certain ion in the first sequence;

[0013] Obtaining modification qualities of different variable modifications, and accurately aligning the rough alignment sequences according to the mapping relationship and the modification qualities of the different variable modifications to obtain accurate alignment sequences;

[0014] Establishing a directed acyclic graph according to the precise alignment sequence;

[0015] parsing the best matching path from the directed acyclic graph, determining the modification position of the variable modification, and obtaining a candidate protein variant;

[0016] The candidate protein variants are screened by a scoring function to obtain the best protein variant corresponding to the protein mass spectrum, and the structure of the best protein variant is used as the structure of the protein to be identified.

[0017] Optionally, comparing the masses of ions in the first sequence and the second sequence to obtain the number of ions in the first sequence whose ion masses are within an error range, calculating the ion matching degree between the first sequence and the second sequence according to the number of ions and a matching degree formula, and selecting n first sequences with top ion matching degrees as the third sequence includes:

[0018] Acquire the number of first ions of the first sequence and the number of second ions of the second sequence;

[0019] Calculate the ion matching degree of each first sequence with the second sequence according to the first ion quantity, the second ion quantity, the quantity of ions of the first sequence whose ion mass is within the error range, and a preset matching degree formula;

[0020] The ion matching degrees of all the first sequences and the second sequences are calculated, and the first n sequences with the highest ion matching degrees are selected as the third sequence.

[0021] Optionally, the matching degree formula is:

[0022] ;

[0023] Wherein, m represents the first sequence, k represents the second sequence, It represents the total number of matching ions in sequence m and sequence k that do not have modified mass shift. represents the total number of ions in sequence m, represents the total number of ions in sequence k, It represents the ion matching degree between the first sequence m and the second sequence k.

[0024] Optionally, obtaining a corresponding peptide tag according to the difference between the masses of mass spectrum peak ions in the second sequence, matching the peptide tag with multiple first sequences to obtain the longest peptide tag length that each first sequence can match, sorting the first sequences according to the length of the longest peptide tag that can be matched, selecting the first m first sequences as the fourth sequence, and combining the third sequence and the fourth sequence as a candidate sequence includes:

[0025] Subtracting the masses of the mass spectrum peak ions in the second sequence in pairs to obtain the corresponding peptide tag masses, arranging and combining the peptide tag masses to obtain different combined masses, and matching the combined masses with the amino acid masses to obtain the corresponding peptide tags;

[0026] Match the peptide tag with the first sequence, traverse the first sequence, find the longest length of the peptide tag that can be matched from the first sequence, sort the longest lengths of the peptide tags by length, and select the first sequences corresponding to the first m peptide tags with the longest lengths as preliminary sequences;

[0027] Determining whether the difference between the protein sequence mass of each of the first sequences in the preliminary sequences and the mass spectrometry precursor mass of the second sequence is within a preset range;

[0028] If the difference between the protein sequence mass of each of the first sequences in the preliminary sequence and the mass spectrometry precursor mass of the second sequence is not within a preset range, the first sequence is deleted from the preliminary sequence to obtain a fourth sequence, and the third sequence and the fourth sequence are combined as a candidate sequence.

[0029] Optionally, the step of roughly aligning the second sequence with the candidate sequence based on a continuous wavelet dynamic time warping algorithm to obtain a mapping relationship between ions of the second sequence and ions of the first sequence, and obtaining a roughly aligned sequence according to the mapping relationship, the second sequence, and the first sequence includes:

[0030] Calculate the second sequence and the first sequence respectively through the wavelet transformation function to obtain a second wavelet sequence and a first wavelet sequence;

[0031] The first wavelet sequence and the second wavelet sequence are roughly aligned according to the time series according to the dynamic time warping algorithm to obtain the mapping relationship between the theoretical peptide ions and the mass spectrum peak ions;

[0032] A rough alignment sequence is established according to the mapping relationship between theoretical peptide ions and mass spectrum peak ions, the second sequence and the first sequence.

[0033] Optionally, the acquiring modification qualities of different variable modifications, and accurately aligning the rough alignment sequence according to the mapping relationship and the modification qualities of the different variable modifications to obtain the accurate alignment sequence comprises:

[0034] According to the mapping relationship, an aligned ion mass pair without mass difference and an ion mass pair with mass difference are obtained;

[0035] Calculating the ion mass difference of the ion mass pair with the mass difference, and determining whether the ion mass difference is equal to the modification mass of a single variable modification or the sum of the modification masses of multiple variable modifications;

[0036] If they are equal, the ion mass pair with a mass difference whose ion mass difference is equal to the modification mass of a single variable modification or the sum of the modification masses of multiple variable modifications is taken as an extended ion pair;

[0037] If not, the ion mass pairs with mass difference whose ion mass difference is not equal to the modification mass of a single variable modification or the sum of the modification masses of multiple variable modifications are used as the filtering ion pairs;

[0038] The rough alignment sequence is expanded and filtered according to the expanded ion pairs and the filtered ion pairs to obtain a precise alignment sequence.

[0039] Optionally, parsing the best matching path from the directed acyclic graph to determine the modification position of the variable modification to obtain the candidate protein variant comprises:

[0040] Taking each node of the directed acyclic graph as a starting point, traversing the path of each matching node, and obtaining the number of nodes on the path, obtaining the path with the largest number of nodes as the optimal matching path;

[0041] Based on the best matching path and the modification positions of the variable modifications, candidate protein variants are obtained.

[0042] A protein mass spectrometry identification system, comprising:

[0043] A first construction module is used to construct the first sequence composed of theoretical peptide ions according to a protein reference sequence, and the number of the first sequence is multiple;

[0044] The second construction module is used to obtain a mass spectrum file corresponding to the protein to be identified, and for each mass spectrum in the mass spectrum file, construct a second sequence consisting of mass spectrum peak ions;

[0045] A first screening module is used to compare the masses of ions in the first sequence and the second sequence, obtain the number of ions in the first sequence whose ion masses are within an error range, calculate the ion matching degree between the first sequence and the second sequence according to the number of ions and a matching degree formula, and select the first n sequences with the highest ion matching degrees as the third sequence;

[0046] a second screening module, for obtaining a corresponding peptide tag according to the difference between the masses of the mass spectrum peak ions in the second sequence, matching the peptide tag with the plurality of the first sequences to obtain the longest peptide tag length that each of the first sequences can match, sorting the peptide tag lengths by length, selecting the first sequences corresponding to the first m peptide tags with the longest lengths as the fourth sequence, and combining the third sequence and the fourth sequence as a candidate sequence;

[0047] a rough alignment module, configured to roughly align the second sequence with the candidate sequence based on a continuous wavelet dynamic time warping algorithm, obtain a mapping relationship between ions of the second sequence and ions of the first sequence, and obtain a rough alignment sequence according to the mapping relationship, the second sequence and the first sequence, wherein the mapping relationship is that the ion mass of one or more ions is equal to the ion mass of a certain ion in the first sequence within an error range or there is a mass difference;

[0048] A precise alignment module, used for obtaining the modification qualities of different variable modifications, and accurately aligning the rough alignment sequences according to the mapping relationship and the modification qualities of the different variable modifications to obtain a precise alignment sequence;

[0049] A generating module, used for establishing a directed acyclic graph according to the precise alignment sequence;

[0050] A matching module, used for parsing the best matching path from the directed acyclic graph, determining the modification position of the variable modification, and obtaining a candidate protein variant;

[0051] The scoring module is used to screen the candidate protein variants through a scoring function to obtain the best protein variant corresponding to the protein mass spectrum, and use the structure of the best protein variant as the structure of the protein to be identified.

[0052] The present invention provides a terminal device, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor loads and executes the computer program, the protein mass spectrometry identification method is adopted.

[0053] The present invention provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium. When the computer program is loaded and executed by a processor, the protein mass spectrometry identification method is adopted.

[0054] The beneficial effect of the present invention is as follows: the first sequence composed of theoretical peptide ions and the second sequence composed of mass spectrum peak ions are firstly screened by the ion matching degree of the first sequence and the second sequence to obtain a third sequence, and then the peptide segment generated by the second sequence is matched with the first sequence, and the peptide segment is screened according to the length of the matched peptide segment to obtain a fourth sequence, and the third sequence and the fourth sequence constitute a candidate sequence, and then the second sequence and the fourth sequence are roughly aligned and precisely aligned to generate a directed acyclic graph, and candidate proteins are obtained according to the directed acyclic graph, and the candidate proteins are screened by a scoring function to obtain an identification result. Firstly, the first sequence is screened and then identified, which can reduce the amount of data processing and improve the identification speed. Then, in the identification process, the position of the modifiable ion is accurately located by rough alignment and precise alignment, thereby improving the accuracy of protein mass spectrometry identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic diagram of a protein mass spectrometry identification method of the present invention;

[0056] Figure 2 This figure compares the protein and prsm results of protein mass spectrometry identification of the 2DLC_H2A_ms2.msalign data set using the wtop method of the present invention and the topmg method under the same input parameters.

[0057] Figure 3 The protein and prsm results of protein mass spectrometry identification of the 2DLC_H31_ms2.msalign data set using the wtop method of the present invention and the topmg method under the same input parameters are compared.

[0058] Figure 4 The protein and prsm results of protein mass spectrometry identification of the 2DLC_H4_ms2.msalign data set were compared between the wtop method of the present invention and the topmg method under the same input parameters.

[0059] Figure 5 The figure shows the comparison of the peak matching results of the 2DLC_H2A_ms2.msalign data set using the wtop method and the topmg method of the present invention with the same input parameters.

[0060] Figure 6The figure shows the comparison of the peak matching results of the 2DLC_H31_ms2.msalign data set with the same input parameters by the wtop method and the topmg method of the present invention.

[0061] Figure 7 The matching peak results of the 2DLC_H4_ms2.msalign data set are compared with the wtop method and the topmg method of the present invention under the same input parameters.

[0062] Figure 8 The figure shows the comparison of the running time required for the programs on the same device when the wtop method and the topmg method of the present invention have the same input parameters. DETAILED DESCRIPTION Example 1

[0063] A protein mass spectrometry identification method, such as Figure 1 As shown, including:

[0064] S100. Constructing a first sequence composed of theoretical peptide ions according to a protein reference sequence, wherein the number of the first sequence is multiple.

[0065] Specifically, the protein reference sequence is obtained from a protein reference database. In this embodiment, a human protein reference sequence library is selected. 2DLC_H2A.mzXML, 2DLC_H31.mzXML, and 2DLC_H4.mzXML data sets can be downloaded from the PeptideAtlas website, and deconvoluted using the topfd software to obtain 2DLC_H2A_ms2.msalign, 2DLC_H31_ms2.msalign, 2DLC_H4_ms2.msalign, and secondary mass spectrometry data, respectively. Protein reference sequence data, amino acid modification site data, variable modification PTMs, protein structure variation and other related data can be obtained on the Uniport website.

[0066] Theoretical peptide ions refer to the ion mass of the peptide that may be produced theoretically calculated based on the amino acid sequence of the protein. Each ion has an ion mass, and the first sequence is a sequence composed of ion masses.

[0067] S110, obtaining a mass spectrum file corresponding to the protein to be identified, and constructing a second sequence consisting of mass spectrum peak ions for each mass spectrum in the mass spectrum file.

[0068] Specifically, a protein mass spectrum file is a file format specifically used to store protein mass spectrum data. The protein to be identified is analyzed by a mass spectrometer to obtain a protein mass spectrum file. The mass spectrum is parsed into mass spectrum peaks, each of which represents a specific ion. The second sequence is a sequence composed of the mass of the mass spectrum peak ions.

[0069] S120, comparing the masses of ions in the first sequence and the second sequence, obtaining the number of ions in the first sequence whose ion masses are within an error range, calculating the ion matching degree between the first sequence and the second sequence according to the number of ions and a matching degree formula, and selecting the top n first sequences with the highest ion matching degrees as the third sequence.

[0070] Specifically, the error range is set according to the needs, usually 15ppm. Compare the masses of the ions in the first sequence and the second sequence, find the number of ions in the first sequence that are equal to the ion mass of the second sequence within the error range, and calculate the ion matching degree. The ion matching degree describes the percentage of the number of ions with equal mass in the second sequence and the first sequence to the sum of the number of ions in the first sequence and the second sequence. The number n can be set according to the needs. In this embodiment, n is 100. That is, the first sequence with the top 100 ion matching degrees is selected to form the third sequence.

[0071] S130. Obtain corresponding peptide tags according to the difference between the masses of mass spectrum peak ions in the second sequence, match the peptide tags with multiple first sequences, obtain the longest peptide tag length that can be matched by each first sequence, sort the peptide tag lengths by length, select the first sequences corresponding to the first m peptide tags with the longest length as the fourth sequence, and combine the third sequence and the fourth sequence as a candidate sequence.

[0072] Specifically, the mass peak ions in the second sequence are subtracted from each other to obtain the mass difference. The amino acid composition of the peptide tag is arranged and combined by the mass difference, and then the peptide tag is used to match in the database to obtain the corresponding peptide tag, and several peptide tags of different lengths can be obtained. Then, multiple peptides in the second sequence are matched with the first sequence to obtain the longest peptide length that can be matched in the first sequence. Until all the first sequences are matched, the first sequences are sorted according to the peptide length of the longest peptide that can be matched, and the first m first sequences are selected as the fourth sequence. For example, there are 3 peptides in the second sequence, with lengths of 5, 4, and 3 respectively. When the second sequence matches the first sequence of A, the length of the longest peptide matched is 4, and when the second sequence matches the first sequence of B, the length of the longest peptide matched is 5. When the second sequence matches the first sequence of C, the length of the longest peptide matched is 3, then the sorting is B first sequence, A first sequence and C first sequence. Assuming the first two are taken, it is B first sequence and A first sequence.

[0073] S140. Based on the continuous wavelet dynamic time warping algorithm, the second sequence is roughly aligned with the candidate sequence to obtain a mapping relationship between the ions of the second sequence and the ions of the first sequence. A rough alignment sequence is obtained according to the mapping relationship, the second sequence and the first sequence. The mapping relationship is that the ion mass of one or more ions in the second sequence is equal to the ion mass of a certain ion in the first sequence within an error range or there is a mass difference.

[0074] Specifically, the second sequence is roughly aligned with the candidate sequence by establishing a mapping relationship between the ion masses according to the ion masses of the second sequence and the ion masses of the first sequence, and preliminarily determining the ions that can be matched in the second sequence.

[0075] S150, obtaining modification qualities of different variable modifications, and accurately aligning the rough alignment sequence according to the mapping relationship and the modification qualities of the different variable modifications to obtain an accurately aligned sequence.

[0076] Specifically, during rough alignment, the ion masses are considered equal, or there is a mass difference in the ion masses. The mass difference is not a fixed value, and there can be multiple different mass differences. The difference is a single modified mass or the sum of multiple modified masses. Therefore, during rough alignment, there may be three situations in which the ion masses of the first sequence and the second sequence are one-to-one, many-to-one, and one-to-many.

[0077] The precise alignment sequence is based on the rough alignment sequence, taking into account the quality of variable modifications, and then expanded and filtered to make the mapping relationship more accurate.

[0078] Variable modifications refer to the variable modifications that occur to proteins or peptides in the sample during sample preparation or the experimental process. These modifications may be naturally occurring biochemical modifications or artificially introduced modifications.

[0079] S160. Establish a directed acyclic graph according to the precise alignment sequence.

[0080] S170. Analyze the optimal matching path from the directed acyclic graph, determine the modification position of the variable modification, and obtain the candidate protein variant.

[0081] Specifically, in the directed acyclic graph, each node represents a mass of a mass spectrometry ion, and the weight of the edge represents the mass offset of the variable modification. Each path in the directed acyclic graph corresponds to a protein variant. By analyzing the optimal matching path, the modification position of the variable modification is determined, and the ions of the second sequence are matched one by one with the ions of the first sequence, and the best protein is obtained as a candidate protein variant.

[0082] S180. Screening candidate protein variants by using a scoring function to obtain the best protein variant corresponding to the protein mass spectrum, and using the structure of the best protein variant as the structure of the protein to be identified.

[0083] Specifically, since the above operation is performed on a first sequence, a candidate protein variant can be obtained, but there are multiple first sequences, so the number of protein variants is also multiple. The candidate protein variants are screened by the scoring function, and the candidate protein variant with the highest score is selected as the best protein variant.

[0084] In one implementation of this embodiment, step S120 compares the masses of ions in the first sequence and the second sequence, obtains the number of ions in the first sequence whose ion masses are within an error range, calculates the ion matching degree between the first sequence and the second sequence according to the number of ions and the matching degree formula, and selects the top n first sequences with the highest ion matching degrees as the third sequence, including:

[0085] S200, obtaining the number of first ions of the first sequence and the number of second ions of the second sequence.

[0086] Specifically, the number of first ions in the first sequence refers to the number of all theoretical peptide ions in the first sequence, and the number of second ions refers to the number of all mass spectrum peak ions in the second sequence.

[0087] S210, calculating the ion matching degree between each first sequence and the second sequence according to the first ion quantity, the second ion quantity, the quantity of ions of the first sequence whose ion mass is within an error range, and a preset matching degree formula.

[0088] Specifically, the number of ions of the first sequence whose ion mass is within the error range is that the difference between the ion mass of the first sequence and the ion mass of the second sequence is within the error range, and the error range is usually set to 15 ppm.

[0089] The matching formula is: ;

[0090] Among them, m represents the first sequence, k represents the second sequence, It represents the total number of matching ions in sequence m and sequence k that do not have modified mass shift. represents the total number of ions in sequence m, represents the total number of ions in sequence k, It represents the ion matching degree between the first sequence m and the second sequence k.

[0091] S220, calculating the ion matching degrees of all the first sequences and the second sequences, and selecting n first sequences with the highest ion matching degrees as the third sequence.

[0092] Each match between the first sequence and the second sequence will result in an ion matching degree. After the second sequence is matched with all the first sequences, the first n first sequences are selected from high to low according to the ion matching degrees, and the selected first n first sequences form the third sequence.

[0093] In one implementation of this embodiment, step S120 is to obtain a corresponding peptide tag according to the difference between the masses of the mass spectrum peak ions in the second sequence, match the peptide tag with multiple first sequences, obtain the longest peptide tag length that each first sequence can match, sort the first sequences according to the length of the longest peptide tag that can be matched, select the first m first sequences as the fourth sequence, and combine the third sequence and the fourth sequence as a candidate sequence, including:

[0094] S300, subtracting the masses of the mass spectrum peak ions in the second sequence in pairs to obtain the corresponding peptide tag masses, arranging and combining the peptide tag masses to obtain different combination masses, matching the combination masses with the amino acid masses to obtain the corresponding peptide tag.

[0095] Specifically, the masses of the mass peak ions in the second sequence are subtracted pairwise, and each mass peak ion in the second sequence will be subtracted from the mass peak ions except for itself to obtain a number of subtracted masses, which are then arranged and combined to match the amino acid masses in the existing database to obtain the corresponding peptide tag.

[0096] S310, matching the peptide tag with the first sequence, traversing the first sequence, finding the longest length of the peptide tag that can be matched from the first sequence, sorting the longest lengths of the peptide tags by length, and selecting the first sequence corresponding to the first m peptide tags with the longest length as the preliminary sequence.

[0097] Specifically, multiple peptides obtained by subtracting each other's mass spectrum peak ions in the second sequence are searched and matched in the first sequence to obtain the longest peptide length that can be matched in the first sequence. At this time, the second sequence is matched with one first sequence, and the second sequence continues to match with the remaining first sequences to obtain the longest peptide length. Then, the first sequences are sorted according to the longest peptide length, and the first m first sequences are selected as preliminary sequences.

[0098] S320, determining whether the difference between the protein sequence mass of each first sequence in the preliminary sequence and the mass spectrometry precursor mass of the second sequence is within a preset range.

[0099] S330. If the difference between the protein sequence mass of each first sequence in the preliminary sequence and the mass spectrometry precursor mass of the second sequence is not within a preset range, the first sequence is deleted from the preliminary sequence to obtain a fourth sequence, and the third sequence and the fourth sequence are combined as a candidate sequence.

[0100] Specifically, the preset range can set the difference range by itself, usually 15ppm. Although the first sequence is preliminarily screened according to the length of the peptide segment, there is a large difference between the ion mass of the matched first sequence and the ion mass of the second sequence. This type of first sequence is not very useful for identifying the protein mass spectrum, so it is deleted from the preparatory sequence, and the remaining first sequence in the preparatory sequence is used as the fourth sequence. Protein sequence mass refers to the mass of the complete protein sequence, and the protein precursor mass can be obtained from the mass spectrometry file. Protein precursor mass refers to the original mass of the molecule (or ion) to be analyzed in mass spectrometry analysis.

[0101] The mass spectrometry precursor mass is the mass of the entire second sequence, including the theoretical peptide mass and the masses of modifications present.

[0102] The mass spectrometry precursor mass (PRECURSORMASS) is used in combination with the theoretical peptide mass (PeptideFragmentMass) and variable modification mass combination (PTMsMass) constructed from the protein reference sequence to detect the most likely protein form of each protein reference sequence, ensuring that the final protein variant mass is closest to the mass spectrometry precursor mass and PPM < 15. This mass relationship is used to supplement the candidate protein sequence. The mass relationship formula is as follows:

[0103] ,in, is the precursor mass, is the theoretical peptide mass, A combination of variable modification qualities.

[0104] The first sequence with a huge amount of data is processed and a part of the sequences is screened out as candidate sequences, thereby reducing the amount of data processing.

[0105] In one implementation of this embodiment, step S140 is to roughly align the second sequence with the candidate sequence based on the continuous wavelet dynamic time warping algorithm to obtain a mapping relationship between ions of the second sequence and ions of the first sequence, and obtaining a roughly aligned sequence according to the mapping relationship, the second sequence and the first sequence includes:

[0106] S400, respectively calculating the second sequence and the first sequence through a wavelet transformation function to obtain a second wavelet sequence and a first wavelet sequence.

[0107] S410 , roughly aligning the first wavelet sequence and the second wavelet sequence according to the time series according to a dynamic time warping algorithm, and obtaining a mapping relationship between theoretical peptide ions and mass spectrum peak ions.

[0108] S420, establishing a rough alignment sequence according to the mapping relationship between theoretical peptide ions and mass spectrum peak ions, the second sequence and the first sequence.

[0109] Specifically, the top-down protein mass spectrometry identification technology involves a large amount of data and computational problems. In each step of protein mass spectrometry identification, the main problem to be solved is how to align and compare the first sequence and the second sequence, and then analyze the protein variant corresponding to the first sequence in the sequence alignment as the top-down protein mass spectrometry identification result. In the analysis process of this method, the candidate sequence and the second sequence are respectively calculated according to the wavelet transform mother function by continuous wavelet transform, and the corresponding first wavelet sequence and the second wavelet sequence are respectively calculated, and the wavelet sequence multi-level transformation is used to highlight the characteristic points of the wavelet sequence, that is, each ion mass in the sequence. The wavelet sequence alignment uses a dynamic time warping algorithm. The dynamic time warping algorithm can be used to preliminarily quickly regularize and align the characteristic values ​​of the candidate sequence with the characteristic values ​​(ion mass) of the first sequence according to the characteristics of time series alignment. Since the second sequence contains noise data and two types of ion masses of the protein N-terminus and C-terminus, as well as the ion mass with variable mass modification offset, the obtained sequence is a rough alignment. Based on the protein mass spectrometry identification method of continuous wavelet dynamic time warping, the formula of continuous wavelet transform is as follows:

[0110] , where t represents the index of a single mass in the sequence, i represents the length of the sequence, W(t) represents a one-dimensional sequence of length i, and j represents the translation amount. It represents the continuous wavelet transform mother function in the time domain and frequency domain. The function uses , It represents the continuous wavelet transform result under length i and offset j.

[0111] After rough alignment, the mapping relationship between the mass spectral peak ions in the second sequence and the theoretical peptide ions of the candidate sequence was determined. The mapping relationship was that the ion mass of one or more ions in the second sequence was equal to the ion mass of an ion in the first sequence within the error range or there was a mass difference (a single modification mass or the sum of the modification masses of multiple modifications). Based on the mapping relationship, a rough alignment sequence was obtained.

[0112] In one implementation of this embodiment, step S150 is to obtain the modification qualities of different variable modifications, and accurately align the rough alignment sequence according to the mapping relationship and the modification qualities of different variable modifications, so as to obtain the accurate alignment sequence, including:

[0113] S500 , obtaining aligned ion mass pairs without mass difference and ion mass pairs with mass difference according to the mapping relationship.

[0114] Specifically, an aligned ion mass pair is a pair of ions formed by the equal ion mass of the second sequence ions and the first sequence ions. For example, the mass of ion A in the second sequence is X, and the mass of ion B in the first sequence is Y. When X=Y+C, A and B are called an ion mass pair, and C is a constant. When X=Y, A and B become an aligned ion mass pair.

[0115] S510, calculating the ion mass difference of the ion mass pairs with mass difference, and determining whether the ion mass difference is equal to the modification mass of a single variable modification or the sum of the modification masses of multiple variable modifications.

[0116] Specifically, the variable modification is manually input, there are multiple variable modifications, and the modification mass of each variable modification is different. In the aligned ion mass pairs, it is further expanded to determine whether the second sequence ions corresponding to the ion mass pairs can be searched from the first sequence to find ions whose mass difference with the second sequence ions is the ion mass of a single variable modification or the sum of multiple variable modification masses. If it can be searched, step S520 is executed.

[0117] When the ion mass pairs are roughly aligned, a mapping relationship is established between the ions of the second sequence and the ions of the first sequence, and the ion pairs with a mapping relationship and a mass difference are regarded as ion mass pairs, that is, the ion masses of the ions of the first sequence and the second sequence that have established a mapping relationship have a mass difference.

[0118] S520, if they are equal, taking the ion mass pair with a mass difference whose ion mass difference is equal to the modification mass of a single variable modification or the sum of the modification masses of multiple variable modifications as the extended ion pair.

[0119] S530, if not equal, taking the ion mass pairs with mass difference whose ion mass difference is not equal to the modification mass of a single variable modification or the sum of the modification masses of multiple variable modifications as the filtering ion pairs.

[0120] Specifically, due to the continuous wavelet transform used in the rough alignment, during the mass alignment, there may be cases where the difference between the ion mass of the second sequence and the ion mass of the first sequence is 0 within the error range, the mass difference is the modified mass, and the mass difference is not the modified mass. The aligned ion mass pairs whose mass difference is not the variable modified mass will be deleted.

[0121] S540, expanding and filtering the roughly aligned sequence according to the extended ion pairs and the filtered ion pairs to obtain a precisely aligned sequence.

[0122] Specifically, based on the rough alignment sequence, an expansion and filtering algorithm for matching ion mass pairs is designed. The expansion algorithm uses the first sequence as the reference sequence and tries to find all matching ion masses with variable modification mass offsets in the second sequence. The filtering algorithm filters matching ion mass pairs that do not meet the variable modification mass offset according to the input variable modification mass and the current protein variant form. The specific steps are as follows: first, a mass range is calculated according to the input variable modification mass parameters based on the feature points of the second sequence. Secondly, all the second sequence feature points within this range are calculated and filtered with the feature points of the first sequence respectively. If there is a variable modification mass offset between the current second sequence feature point and the feature point (ion mass) of the first sequence and there is no mapping relationship, a new mapping relationship is established between the feature point of the second sequence and the feature point of the current first sequence. Finally, the feature points of the first sequence and the feature points of the second sequence will form a one-to-one or one-to-many precise mapping relationship. Compared with the mapping relationship, the precise mapping relationship takes into account the variable modification mass, so that the mapping relationship between the ions in the second sequence and the ions in the first sequence is more accurate.

[0123] In one implementation of this embodiment, step S170 is to parse the optimal matching path from the directed acyclic graph, determine the modification position of the variable modification, and obtain the candidate protein variants including:

[0124] S600, taking each node of the directed acyclic graph as a starting point, traversing the path of each node, and obtaining the number of nodes on the path, and obtaining the path with the largest number of nodes as the optimal matching path.

[0125] S610: Obtain candidate protein variants based on the optimal matching path and the modification positions of the variable modifications.

[0126] Specifically, a directed acyclic graph is established based on the alignment sequence to characterize the alignment mapping relationship between the current candidate first sequence and the second sequence. In the directed acyclic graph, each node represents a mass of a mass spectrometer ion, and the weight of the edge represents the mass offset of the variable modification. Each path in the directed acyclic graph corresponds to a protein variant. By obtaining and parsing the optimal matching path, the mass and position of the variable modification are determined to obtain the final protein variant. The optimal matching path is calculated based on the weights and nodes of the edges in the directed acyclic graph. The specific steps are as follows: first, each node of the graph is used as the starting node of the current path, and all paths of the starting node are traversed. Secondly, while traversing each path, the nodes on each path are counted, and the path with the most nodes is selected as the optimal matching path of the current directed acyclic graph. Finally, the optimal matching path is parsed in combination with the input variable modification information, and the protein variant with the most matching ion mass pairs that is finally identified is converted into a candidate protein variant.

[0127] The candidate protein variants are screened by the scoring function to obtain the best protein variant corresponding to the protein mass spectrum. The structure of the best protein variant is used as the structure of the protein to be identified, including:

[0128] The alignment sequence scoring function was defined based on the matching characteristics of continuous ions and complementary ions, and the protein mass spectrometry identification results were output after sorting.

[0129] Continuous ion mass matching means that there are two mass spectrometry ion masses, and the peptides corresponding to the matched theoretical peptide ion masses differ by only one amino acid, and the amino acid sequence of the short peptide is the prefix of the amino acid sequence of the long peptide. Complementary ion mass matching means that there are two mass spectrometry ion masses, and the peptides corresponding to the matched theoretical peptide ion masses are the prefix and suffix of the protein sequence, respectively. The two sequences are spliced ​​to form a complete protein sequence. The alignment sequence scoring function is defined based on the matching characteristics of continuous ions and complementary ions. The scoring function formula is:

[0130] , It represents the score of each node in the N-terminal matching path sequence i. It represents the score of each node in the C-terminal matching path sequence j, where the N-terminal is the starting segment of the protein sequence and the C-terminal is the ending segment of the protein sequence. The truncated lengths of the N-terminal and C-terminal of the protein variant sequence and the removal of 10, It represents the complementary node coefficients in i and j. represents consecutive matching nodes in vector i, It represents the number of nodes in i. represents the number of nodes in j, It represents the matching score of protein mass spectrometry matching paths i and j. After sorting by matching score, the candidate sequence with the highest score for each second sequence is obtained, and converted into protein variants that can be identified by the corresponding mass spectrometry, and the identification results are output. Example 2

[0131] In order to verify the performance of the method wtop of the present invention, the toppic method in top-down protein mass spectrometry identification was selected for comparison. The secondary mass spectrometry files after deconvolution of 2DLC_H2A.mzXML, 2DLC_H31.mzXML and 2DLC_H4.mzXML were selected as mass spectrometry data sets, the protein reference sequence file was the human protein sequence library (human+decoy), and the variable modifications were the 5 most commonly used protein modifications in human proteins, as shown in Table 1.

[0132] Table 1 Variable modifications

[0133]

[0134] The parameter settings of the wtop and topmg methods are shown in Table 2. The number of protein identifications, the number of protein variant mass spectra matches (PrSMs), and the number of effective mass spectrometry identifications of the corresponding data sets were calculated, and the results were compared and analyzed. After false positive analysis of all results, Figure 2-Figure 4 shown.

[0135] Table 2 Parameter settings

[0136] Example 3

[0137] Quantitative analysis and comparison of PrSMs of protein mass spectrometry identification results based on continuous wavelet dynamic time warping protein mass spectrometry identification method.

[0138] from Figure 2 It can be seen that with the same parameters and input data sets, the number of proteins and PrSMs that the wtop method can identify on the corresponding data sets are greater than those of the topmg method.

[0139] Based on the comparison of the two-dimensional bar graphs, the following criteria were used for identification, including the proportion of reported PrSMs in the dataset (PrSMs / coveragerate1) and the proportion of unique PrSMs (UniquePrSMs / coveragerate3). It can be seen from Tables 3, 4, and 5 that the number of unique PrSMs associated with the dataset reported by the wtop method on the H2A, H31, and H4 datasets is higher than the number of unique PrSMs associated with the dataset reported by topmg. It can be seen from Tables 3-5 that the number of PrSMs associated with the corresponding dataset reported by the wtop method is higher than the number of PrSMs associated with the corresponding dataset reported by the topmg method.

[0140] In summary, from Figure 2-Figure 5 Tables 3-5 show that in top-down protein mass spectrometry identification, when wtop has the same parameters and input data set as the toppic method, the PrSMs results of protein mass spectrometry identification are superior to the topmg identification method in terms of quantity and accuracy.

[0141] Table 3 Comparison of mass spectrometry matching numbers of protein variants in the H2A dataset

[0142]

[0143] Table 4 Comparison of the number of mass spectrometric matches of protein variants in the H31 dataset

[0144]

[0145] Table 5 Comparison of the number of mass spectrometric matches of protein variants in the H4 dataset

[0146] Example 4

[0147] Analysis and comparison of the mass quantity of PrSMs matching ions in the protein mass spectrometry identification results based on the continuous wavelet dynamic time rule algorithm.

[0148] The performance of the wtop method and other methods was further verified using ion matching histograms. Figure 5-Figure 7 The horizontal axis represents the interval of the number of matching ions, and the vertical axis represents the number of PrSMs corresponding to each interval of the number of matching ions. Figure 5 The comparison between the number of matching ions of the wtop method and the topmg method on the H2A dataset is shown. In order to eliminate the chance, Figure 6 and Figure 7The comparison of the number of matching ions between the two methods on the H31 dataset and the H4 dataset is shown respectively. It can be seen that in the H2A and H31 datasets, the total number of PrSMs of wtop under different matching ion numbers is higher than the total number of PrSMs of the topmg method. In the H4 dataset, when the number of matching ions is less than 60, the total number of PrSMs of wtop is higher than the total number of PrSMs of the topmg method. It can be seen that the wtop method can report more PrSMs of matching ions than the topmg method. Example 5

[0149] Analysis and comparison of running time of protein mass spectrometry identification methods based on continuous wavelet dynamic time rules.

[0150] On the same device and with the same conditions of 4 threads, the running time of the wtop method is compared with that of topmg on three data sets. The comparison results are shown in Figure 8 As shown in the figure, in these three data sets, it can be found that the running time of the topmg program is higher than that of the wtop program in all three data sets. In the H2A and H3.1 data sets, the running time required by topmg is 74 times and 172 times that of wtop respectively. In the HistoneH4 data set, the running time of topmg is also 90 times that of wtop.

[0151] A protein mass spectrometry identification system, comprising:

[0152] A first construction module is used to construct a first sequence composed of theoretical peptide ions according to a protein reference sequence, and the number of the first sequence is multiple;

[0153] The second construction module is used to obtain a mass spectrum file corresponding to the protein to be identified, and for each mass spectrum in the mass spectrum file, construct a second sequence consisting of mass spectrum peak ions;

[0154] The first screening module is used to compare the masses of ions in the first sequence and the second sequence, obtain the number of ions in the first sequence whose ion masses are within the error range, calculate the ion matching degree between the first sequence and the second sequence according to the number of ions and the matching degree formula, and select the first n sequences with the highest ion matching degree as the third sequence;

[0155] The second screening module is used to obtain the corresponding peptide tag according to the difference between the masses of the mass spectrum peak ions in the second sequence, match the peptide tag with the multiple first sequences, obtain the longest peptide tag length that can be matched by each first sequence, sort the peptide tag lengths according to the length, select the first sequences corresponding to the first m peptide tags with the longest length as the fourth sequence, and combine the third sequence and the fourth sequence as a candidate sequence;

[0156] A rough alignment module is used to roughly align the second sequence with the candidate sequence based on a continuous wavelet dynamic time warping algorithm to obtain a mapping relationship between ions of the second sequence and ions of the first sequence, and obtain a rough alignment sequence according to the mapping relationship, the second sequence and the first sequence, wherein the mapping relationship is that the ion mass of one or more ions is equal to the ion mass of a certain ion in the first sequence within an error range or there is a mass difference;

[0157] The precise alignment module is used to obtain the ion masses of different variable modifications, and accurately align the rough alignment sequence according to the mapping relationship and the ion masses of different variable modifications to obtain the precise alignment sequence;

[0158] A generation module, used to build a directed acyclic graph based on the precise alignment sequence;

[0159] A matching module, which is used to parse the best matching path from the directed acyclic graph to obtain candidate protein variants;

[0160] The scoring module is used to screen candidate protein variants through a scoring function to obtain the best protein variant corresponding to the protein mass spectrum, and the structure of the best protein variant is used as the structure of the protein to be identified.

[0161] The embodiment of the present application also discloses a terminal device, including a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, a protein mass spectrometry identification method is adopted.

[0162] The terminal device may be a computer device such as a desktop computer, a laptop computer or a cloud server, and the terminal device includes but is not limited to a processor and a memory. For example, the terminal device may also include input and output devices, a network access device and a bus.

[0163] Among them, the processor can adopt a central processing unit (CPU). Of course, according to actual usage, other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. can also be adopted. The general-purpose processor can adopt a microprocessor or any conventional processor, etc., and this application does not impose any restrictions on this.

[0164] Among them, the memory can be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device, or it can be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (SMC), a secure digital card (SD) or a flash memory card (FC) equipped on the terminal device, and the memory can also be a combination of an internal storage unit and an external storage device of the terminal device. The memory is used to store computer programs and other programs and data required by the terminal device. The memory can also be used to temporarily store data that has been output or is to be output, and this application does not impose any restrictions on this.

[0165] Among them, through this terminal device, a protein mass spectrometry identification method in the above embodiment is stored in the memory of the terminal device, and is loaded and executed on the processor of the terminal device for easy use.

[0166] An embodiment of the present application further discloses a computer-readable storage medium, and the computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, a protein mass spectrometry identification method in the above embodiment is adopted.

[0167] Among them, the computer program can be stored in a computer-readable medium, the computer program includes computer program code, the computer program code can be in the form of source code, object code, executable file or certain middleware, etc. The computer-readable medium includes any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the computer-readable medium includes but is not limited to the above-mentioned components.

[0168] Among them, through this computer-readable storage medium, a protein mass spectrometry identification method in the above embodiment is stored in a computer-readable storage medium, and is loaded and executed on a processor to facilitate the storage and application of the above method.

[0169] A person skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of protection of the present application is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present application as above, which are not provided in detail for the sake of simplicity.

[0170] One or more embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the present application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of the present application should be included in the protection scope of the present application.

Claims

1. A protein mass spectrometry identification method, characterized in that: include: According to the protein reference sequence, construct a first sequence composed of theoretical peptide ions, wherein the number of the first sequence is multiple; Obtaining a mass spectrum file corresponding to the protein to be identified, and constructing a second sequence consisting of mass spectrum peak ions for each mass spectrum in the mass spectrum file; Compare the masses of ions in the first sequence and the second sequence to obtain the number of ions in the first sequence whose ion masses are within the error range, calculate the ion matching degree between the first sequence and the second sequence according to the number of ions and the matching degree formula, and select the ions with the highest ion matching degree. n the first sequence as the third sequence; According to the difference between the masses of the mass spectrum peak ions in the second sequence, the corresponding peptide tag is obtained, and the peptide tag is matched with multiple first sequences to obtain the longest peptide tag length that each first sequence can match, and the first sequences are sorted according to the length of the longest peptide tag that can be matched, and the first sequence is selected. m the first sequence as a fourth sequence, and combining the third sequence and the fourth sequence as a candidate sequence; Based on a continuous wavelet dynamic time warping algorithm, the second sequence is roughly aligned with the candidate sequence to obtain a mapping relationship between ions of the second sequence and ions of the first sequence, and a roughly aligned sequence is obtained according to the mapping relationship, the second sequence and the first sequence, wherein the mapping relationship is that the ion mass of one or more ions in the second sequence is equal to the ion mass of a certain ion in the first sequence within an error range or there is a mass difference; Obtaining modification qualities of different variable modifications, and accurately aligning the rough alignment sequences according to the mapping relationship and the modification qualities of the different variable modifications to obtain accurate alignment sequences; Establishing a directed acyclic graph according to the precise alignment sequence; parsing the best matching path from the directed acyclic graph, determining the modification position of the variable modification, and obtaining a candidate protein variant; The candidate protein variants are screened by a scoring function to obtain the best protein variant corresponding to the protein mass spectrum, and the structure of the best protein variant is used as the structure of the protein to be identified.

2. The protein mass spectrometry identification method according to claim 1, characterized in that: The method comprises comparing the masses of ions in the first sequence and the second sequence, obtaining the number of ions in the first sequence whose ion masses are within an error range, calculating the ion matching degree between the first sequence and the second sequence according to the number of ions and a matching degree formula, and selecting the ions with the highest ion matching degree. n The first sequence as the third sequence includes: Acquire the number of first ions of the first sequence and the number of second ions of the second sequence; Calculate the ion matching degree of each first sequence with the second sequence according to the first ion quantity, the second ion quantity, the quantity of ions of the first sequence whose ion mass is within the error range, and a preset matching degree formula; Calculate the ion matching degree of all the first sequence and the second sequence, and select the ion with the highest matching degree. n The first sequence is used as the third sequence.

3. The protein mass spectrometry identification method according to claim 2, characterized in that: The matching degree formula is: ； in, m represents the first sequence, k represents the second sequence, It represents the sequence m with sequence k The total number of matching ions without modified mass shift in It represents the sequence m The total number of ions in It represents the sequence k The total number of ions in The first sequence m and the second sequence k ion matching.

4. The protein mass spectrometry identification method according to claim 1, characterized in that: According to the difference between the masses of the mass spectrum peak ions in the second sequence, the corresponding peptide tag is obtained, the peptide tag is matched with multiple first sequences to obtain the longest peptide tag length that each first sequence can match, the first sequences are sorted according to the length of the longest peptide tag that can be matched, and the first sequence is selected. m the first sequence as a fourth sequence, and combining the third sequence and the fourth sequence as a candidate sequence comprises: Subtracting the masses of the mass spectrum peak ions in the second sequence in pairs to obtain the corresponding peptide tag masses, arranging and combining the peptide tag masses to obtain different combined masses, and matching the combined masses with the amino acid masses to obtain the corresponding peptide tags; Match the peptide tag with the first sequence, traverse the first sequence, find the longest length of the peptide tag that can be matched from the first sequence, sort the longest lengths of the peptide tags by length, and select the one with the longest length first. m The first sequence corresponding to each peptide tag is used as a preliminary sequence; Determining whether the difference between the protein sequence mass of each of the first sequences in the preliminary sequences and the mass spectrometry precursor mass of the second sequence is within a preset range; If the difference between the protein sequence mass of each of the first sequences in the preliminary sequence and the mass spectrometry precursor mass of the second sequence is not within a preset range, the first sequence is deleted from the preliminary sequence to obtain a fourth sequence, and the third sequence and the fourth sequence are combined as a candidate sequence.

5. The protein mass spectrometry identification method according to claim 1, characterized in that: The step of roughly aligning the second sequence with the candidate sequence based on a continuous wavelet dynamic time warping algorithm to obtain a mapping relationship between ions of the second sequence and ions of the first sequence, and obtaining a roughly aligned sequence according to the mapping relationship, the second sequence and the first sequence comprises: Calculate the second sequence and the first sequence respectively through the wavelet transformation function to obtain a second wavelet sequence and a first wavelet sequence; The first wavelet sequence and the second wavelet sequence are roughly aligned according to the time series according to the dynamic time warping algorithm to obtain the mapping relationship between the theoretical peptide ions and the mass spectrum peak ions; A rough alignment sequence is established according to the mapping relationship between theoretical peptide ions and mass spectrum peak ions, the second sequence and the first sequence.

6. The protein mass spectrometry identification method according to claim 1, characterized in that: The obtaining of modification qualities of different variable modifications, and accurately aligning the rough alignment sequence according to the mapping relationship and the modification qualities of the different variable modifications to obtain an accurate alignment sequence comprises: According to the mapping relationship, an aligned ion mass pair without mass difference and an ion mass pair with mass difference are obtained; Calculating the ion mass difference of the ion mass pair with the mass difference, and determining whether the ion mass difference is equal to the modification mass of a single variable modification or the sum of the modification masses of multiple variable modifications; If they are equal, the ion mass pair with a mass difference whose ion mass difference is equal to the modification mass of a single variable modification or the sum of the modification masses of multiple variable modifications is taken as an extended ion pair; If not, the ion mass pairs with mass difference whose ion mass difference is not equal to the modification mass of a single variable modification or the sum of the modification masses of multiple variable modifications are used as the filtering ion pairs; The rough alignment sequence is expanded and filtered according to the expanded ion pairs and the filtered ion pairs to obtain a precise alignment sequence.

7. The protein mass spectrometry identification method according to claim 6, characterized in that: The step of parsing the optimal matching path from the directed acyclic graph, determining the modification position of the variable modification, and obtaining the candidate protein variant comprises: Taking each node of the directed acyclic graph as a starting point, traversing the path of each node, and obtaining the number of nodes on the path, obtaining the path with the largest number of nodes as the optimal matching path; Based on the best matching path and the modification positions of the variable modifications, candidate protein variants are obtained.

8. A protein mass spectrometry identification system, characterized in that: include: A first construction module is used to construct a first sequence consisting of theoretical peptide ions according to a protein reference sequence, wherein the number of the first sequence is multiple; The second construction module is used to obtain a mass spectrum file corresponding to the protein to be identified, and for each mass spectrum in the mass spectrum file, construct a second sequence consisting of mass spectrum peak ions; The first screening module is used to compare the masses of ions in the first sequence and the second sequence, obtain the number of ions in the first sequence whose ion masses are within the error range, calculate the ion matching degree between the first sequence and the second sequence according to the number of ions and the matching degree formula, and select the ions with the highest ion matching degree. n the first sequence as the third sequence; The second screening module is used to obtain the corresponding peptide tag according to the difference between the masses of the mass spectrum peak ions in the second sequence, match the peptide tag with multiple first sequences, obtain the longest peptide tag length that each first sequence can match, sort the peptide tag lengths according to length, and select the one with the longest length. m The first sequence corresponding to the peptide tag is used as the fourth sequence, and the third sequence and the fourth sequence are combined as a candidate sequence; a rough alignment module, configured to roughly align the second sequence with the candidate sequence based on a continuous wavelet dynamic time warping algorithm, obtain a mapping relationship between ions of the second sequence and ions of the first sequence, and obtain a rough alignment sequence according to the mapping relationship, the second sequence and the first sequence, wherein the mapping relationship is that the ion mass of one or more ions is equal to the ion mass of a certain ion in the first sequence within an error range or there is a mass difference; A precise alignment module, used for obtaining the modification qualities of different variable modifications, and precisely aligning the rough alignment sequences according to the mapping relationship and the modification qualities of the different variable modifications to obtain a precise alignment sequence; A generating module, used for establishing a directed acyclic graph according to the precise alignment sequence; A matching module, used for parsing the best matching path from the directed acyclic graph, determining the modification position of the variable modification, and obtaining a candidate protein variant; The scoring module is used to screen the candidate protein variants through a scoring function to obtain the best protein variant corresponding to the protein mass spectrum, and take the structure of the best protein variant as the structure of the protein to be identified.

9. A terminal device, comprising a memory and a processor, characterized in that: The memory stores a computer program that can be run on the processor, and when the processor loads and executes the computer program, the identification method according to any one of claims 1 to 7 is adopted.

10. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, characterized in that: When the computer program is loaded and executed by a processor, the identification method according to any one of claims 1 to 7 is adopted.

Images