Polypeptide and TCR molecule binding relation prediction method based on large-scale language model

By using the fusion method of bidirectional GRU and RetNet modules in the prediction of binding relationship between peptides and TCR molecules, the problem of poor prediction in the existing technology is solved, and more efficient binding relationship prediction is achieved, assisting clinical immunotherapy and improving treatment success Rate.

CN118942553AActive Publication Date: 2024-11-12CENT SOUTH UNIV

Patent Information

Application Number
CN202411020937.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-12
Estimated Expiration
2044-07-29

AI Technical Summary

Technical Problem

The existing method of binding prediction of peptides to TCR molecules is not ideal for the binding prediction of known peptides, and it is difficult to expand to clinical applications. The effect of the pan-specific tools is poor, so it cannot take into account the integrity and correlation of the sequence during the training process. Failure to effectively mine and visualize the binding preferences of each peptide.

Method used

Using a large-scale language model method, the fusion of bidirectional GRU and RetNet modules is used to model, and the feature coding and relationship prediction of peptides and TCR molecules is performed. The features are extracted using the AAIndex database, and probability prediction is performed by combining the output of bidirectional GRU and RetNet modules.

Benefits of technology

Improve the accuracy and efficiency of prediction of the binding relationship between peptides and TCR molecules, which can assist in target screening in clinical immunotherapy, shorten the treatment cycle, increase the success rate of tumor treatment, and show strong prediction performance in different environments and databases. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118942553A_ABST
    Figure CN118942553A_ABST
Patent Text Reader

Abstract

The invention discloses a polypeptide and TCR molecule binding relationship prediction method based on a large-scale language model, which comprises the following steps: S100, acquiring a polypeptide sequence of a to-be-detected binding relationship and a CDR3 fragment sequence on a TCR molecule, and respectively preprocessing the to-be-detected sequence to obtain a target input sequence; s200, performing feature coding on each amino acid residue in the target input sequence, and outputting a coding matrix of each sequence; s300, respectively inputting the coding matrixes into a bidirectional GRU module in a trained combination relation prediction model, and outputting a global feature matrix of each sequence; s400, merging the global feature matrixes of each pair of sequences, inputting the merged global feature matrixes into a large-scale language model RetNet in a trained combination relation prediction model, and outputting a relation matrix of each pair of sequences; and S500, inputting the relation matrix of each pair of sequences into a binding relation prediction model, and outputting the binding probability of the polypeptide sequence of the binding relation to be detected and the CDR3 fragment sequence on the TCR molecule.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of immunity and tumor treatment technology, and in particular to a method for predicting the binding relationship between a polypeptide and a TCR molecule based on a large-scale language model. Background Art

[0002] TCR is a membrane protein on the surface of T lymphocytes. It guides the immune response of T cells by recognizing antigen fragments (usually presented by antigen-presenting cells). TCR molecules play an important role in the clinical treatment of cancer, vaccine development, and the treatment of autoimmune diseases. Studying the binding mechanism between antigens (peptides) and TCR molecules can promote personalized immunotherapy and innovative vaccine design, and provide new hope for the treatment of some intractable diseases. TCR is encoded by several gene fragments, including α and β chains for most T cells. These gene fragments are combined together during the development of T cells through the process of gene recombination to form a complete TCR gene. This gene recombination process leads to huge genetic diversity. Using deep learning methods to predict the binding of peptides to TCR molecules can help screen targets for cancer treatment, provide assistance for diagnosis, clinical immunotherapy, and the development of new vaccines, can significantly shorten the treatment cycle, and provide possibilities for immunotherapy for cancer patients. At present, the increasing amount of data has brought more opportunities and possibilities for combined prediction tasks.

[0003] At present, there are some methods that are used to predict the binding of peptides to TCR molecules. According to the type of data input that can be predicted, they can be divided into specific tools and pan-specific tools. Specific tools can only establish separate prediction models for peptides that already have binding data, and this model cannot predict the binding of new peptides. Pan-specific tools can predict the binding of new peptides that are not in the training data. These tools characterize the binding of known / unknown peptides to TCR by modeling the binding of a single peptide to TCR or modeling the binding of pan-peptides to TCR. However, these methods are still not ideal for predicting the binding of known peptides, and it is still difficult to expand them to clinical applications; although pan-specific tools can predict the binding of unknown peptides, the effect is very poor. Therefore, new modeling methods are still needed to improve the prediction of peptide-TCR binding and provide assistance for immunotherapy.

[0004] Current methods are still challenging for predicting peptide-TCR binding, mainly in the following aspects: (1) The model cannot take into account the integrity and relevance of the sequence during the training process; the most commonly used modules in current tools are fully connected, convolutional neural network (CNN), autoencoder, self-attention and other modules, but the structure of these models also largely destroys the sequence integrity and relevance of the two input sequences. (2) Although there are many different methods for specificity and pan-specificity, the prediction results are still not ideal for existing epitope data. (3) The current model does not mine and visualize the binding preference of each peptide, which is not conducive to clinical applications (such as TCR modification, mutant peptide vaccines, etc.) and further improvements. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] Based on the above problems, the present invention provides a method for predicting the binding relationship between polypeptides and TCR molecules based on a large-scale language model. The method improves the accuracy and efficiency of binding relationship prediction by feature encoding polypeptides and TCR molecules and adopting a fusion of bidirectional GRU and RetNet modules for modeling.

[0007] (II) Technical solution

[0008] Based on the above technical problems, the present invention provides a method for predicting the binding relationship between a polypeptide and a TCR molecule based on a large-scale language model, comprising the following steps:

[0009] S100, obtaining a polypeptide sequence of a binding relationship to be tested and a CDR3 fragment sequence on a TCR molecule, and preprocessing the sequences to be tested to obtain a target input sequence;

[0010] S200, feature encoding each amino acid residue in the target input sequence, and outputting a coding matrix for each sequence;

[0011] S300, inputting the encoding matrices into the bidirectional GRU modules in the trained combined relationship prediction model respectively, and outputting the global feature matrix of each sequence;

[0012] S400, after merging the corresponding global feature matrices of each sequence, inputting them into the trained large-scale language model RetNet in the combined relationship prediction model, and outputting the relationship matrix of each pair of sequences;

[0013] S500, inputting the relationship matrix of each pair of sequences into a binding relationship prediction model, and outputting the binding probability of the polypeptide sequence of the binding relationship to be tested and the CDR3 fragment sequence on the TCR molecule.

[0014] Further, step S200 includes:

[0015] According to 28 features extracted from the AAIndex database that have been successfully applied in the field of protein binding, the physical and chemical characteristics of each amino acid residue in the target input sequence are encoded.

[0016] Among them, the 28 features that have been successfully applied in the field of protein binding are CHOP780202, CIDH920103, CIDH920105, FAUJ880109, GEIM800106, KANM800102, KLEP840101, KRIW710101, PALJ810107, QIAN880123, RACS770103, RADA880108, ZIMJ68010 4. AURR980120, MUNV940103, NADH010104, FAUJ880111, FINA910104, GEIM800104, NADH010106, LIFS 790101, MEEJ800101, OOBM770102, GUYH850105, ROSM880102, SWER830101, ZIMJ680102, MIYS990104.

[0017] Further, step S300 includes:

[0018] Input the encoding matrix of the peptide into the bidirectional GRU module to obtain the global feature matrix of the peptide;

[0019] The encoding matrix of the CDR3 fragment on the TCR molecule is input into the bidirectional GRU module to obtain the global feature matrix of CDR3.

[0020] Furthermore, step S400 is specifically as follows:

[0021] The global feature matrices of the peptide and CDR3 segments were spliced;

[0022] The concatenated matrix is ​​input into a 24-layer cascaded RetNet module, and the relationship matrix of each pair of sequences is output.

[0023] Furthermore, step S500 is specifically as follows:

[0024] The relationship matrix of each pair of sequences is input into the binding relationship prediction model, and the binding probability of the polypeptide sequence to be tested for the binding relationship and the CDR3 fragment sequence on the TCR molecule is output, specifically:

[0025] Input the relationship matrix into two layers of linear layer and ReLU layer connected in series;

[0026] After another linear layer and a Sigmoid layer, the value is mapped to [0, 1], indicating the binding probability between the input peptide and the TCR molecule.

[0027] The present invention also discloses a device for predicting the binding relationship between a polypeptide and a TCR molecule based on a large-scale language model, which comprises at least one processor; and at least one memory in communication with the processor, wherein:

[0028] The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the above-mentioned method for predicting the binding relationship between polypeptides and TCR molecules based on a large-scale language model, including the following functional modules:

[0029] A data acquisition and preprocessing module, used to execute the step S100;

[0030] A feature encoding module, used to execute step S200;

[0031] A feature representation module, used to execute step S300;

[0032] Combined with the relationship information mining module, used to perform the step S400;

[0033] The prediction module is used to execute the step S500.

[0034] Furthermore, the feature representation module includes a bidirectional GRU module, which can independently learn the context information of each sequence, including the context of the subsequence. The bidirectional GRU module comprehensively learns the information of different subsequences along the sequence direction, ensuring the integrity and sequence of information during the model training process.

[0035] Furthermore, the combined relational information mining module includes a RetNet module, which has 24 layers connected in series. The crucial recursive representation module in the RetNet module is a variant of self-attention, which introduces the D matrix and the GN module. The exponential decay factor Y in the D matrix enables RetNet to retain the semantic influence of sequence proximity during the learning process. In addition, the GN module introduces nonlinearity to facilitate parallel training of modules. In essence, the addition of the RetNet module significantly enhances the model's understanding of the sequence and increases the training speed. Compared with the traditional Transformer position encoding, this matrix is ​​more consistent with sequence processing in terms of position perception and is more suitable for the task of predicting the binding of peptides to TCRs.

[0036] Furthermore, after the RetNet module outputs the relationship matrix of each pair of sequences, it enters the prediction module in sequence through 3 linear layers, 2 ReLU activation layers, and 1 random downsampling. Finally, the result passes through a Sigmoid activation layer to map the value to [0, 1], indicating the binding probability between the input peptide and the TCR molecule.

[0037] The present invention also discloses a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable the computer to execute the above-mentioned method for predicting the binding relationship between polypeptides and TCR molecules based on a large-scale language model.

[0038] (III) Beneficial effects

[0039] The above technical solution of the present invention has the following advantages:

[0040] The prediction method provided by the present invention integrates insights from previous tools and adopts the fusion of bidirectional GRU and RetNet modules for modeling, and has become an excellent tool for predicting the binding of peptides to TCR molecules; a large number of experiments have shown that the method not only shows its advantages in three different environments, but also shows its strong predictive performance in tasks such as characterizing peptide binding preferences, distinguishing true binding compounds from false binding compounds, characterizing T cell clone expansion rates, evaluating the impact of alanine mutations on prediction results, and performance on large external databases; the method can be used to assist in the screening of therapeutic targets in clinical immunotherapy, save target screening time, shorten the treatment cycle, and increase the success rate of tumor treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The features and advantages of the present invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the present invention in any way. In the accompanying drawings:

[0042] Figure 1 is a flow chart of the prediction method of the present invention;

[0043] Figure 2 The results of the present invention and the comparative method under three different test conditions are compared;

[0044] Figure 3 The following is an example of the comparison of the results of the present invention and the comparative method in characterizing the binding preference of the polypeptide;

[0045] Figure 4 Demonstration of the peptide binding ability of the present invention to reproduce CDR3. DETAILED DESCRIPTION

[0046] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0047] The embodiment of the present invention is a method for predicting the binding relationship between a polypeptide and a TCR molecule based on a large-scale language model, comprising the following steps:

[0048] S100, obtaining a polypeptide sequence of a binding relationship to be tested and a CDR3 fragment sequence on a TCR molecule, and preprocessing the sequences to be tested to obtain a target input sequence;

[0049] S200, feature encoding each amino acid residue in the target input sequence, and outputting a coding matrix for each sequence;

[0050] S300, inputting the encoding matrices into the bidirectional GRU modules in the trained combined relationship prediction model respectively, and outputting the global feature matrix of each sequence;

[0051] S400, after merging the corresponding global feature matrices of each sequence, inputting them into the trained large-scale language model RetNet in the combined relationship prediction model, and outputting the relationship matrix of each pair of sequences;

[0052] S500, inputting the relationship matrix of each pair of sequences into a binding relationship prediction model, and outputting the binding probability of the polypeptide sequence of the binding relationship to be tested and the CDR3 fragment sequence on the TCR molecule.

[0053] Wherein, step S200 includes:

[0054] According to 28 features extracted from the AAIndex database that have been successfully applied in the field of protein binding, the physical and chemical characteristics of each amino acid residue in the target input sequence are encoded.

[0055] Among them, the 28 features that have been successfully applied in the field of protein binding are CHOP780202, CIDH920103, CIDH920105, FAUJ880109, GEIM800106, KANM800102, KLEP840101, KRIW710101, PALJ810107, QIAN880123, RACS770103, RADA880108, ZIMJ68010 4. AURR980120, MUNV940103, NADH010104, FAUJ880111, FINA910104, GEIM800104, NADH010106, LIFS790101, MEEJ800101, OOBM770102, GUYH850105, ROSM880102, SWER830101, ZIMJ680102, MIYS990104. See Table 1 for details.

[0056] Table 1

[0057]

[0058] Wherein, step S300 includes:

[0059] Input the encoding matrix of the peptide into the bidirectional GRU module to obtain the global feature matrix of the peptide;

[0060] The encoding matrix of the CDR3 fragment on the TCR molecule is input into the bidirectional GRU module to obtain the global feature matrix of CDR3.

[0061] Wherein, step S400 is specifically as follows:

[0062] The global feature matrices of the peptide and CDR3 segments were spliced;

[0063] The concatenated matrix is ​​input into a 24-layer cascaded RetNet module, and the relationship matrix of each pair of sequences is output.

[0064] Furthermore, step S500 is specifically as follows:

[0065] The relationship matrix of each pair of sequences is input into the binding relationship prediction model, and the binding probability of the polypeptide sequence to be tested for the binding relationship and the CDR3 fragment sequence on the TCR molecule is output, specifically:

[0066] Input the relationship matrix into two layers of linear layer and ReLU layer connected in series;

[0067] After another linear layer and a Sigmoid layer, the value is mapped to [0, 1], indicating the binding probability between the input peptide and the TCR molecule.

[0068] The embodiment of the present invention is a device for predicting the binding relationship between a polypeptide and a TCR molecule based on a large-scale language model, which includes at least one processor; and at least one memory in communication with the processor, wherein:

[0069] The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the above-mentioned method for predicting the binding relationship between polypeptides and TCR molecules based on a large-scale language model, including the following functional modules:

[0070] A data acquisition and preprocessing module, used to execute the step S100;

[0071] A feature encoding module, used to execute step S200;

[0072] A feature representation module, used to execute step S300;

[0073] Combined with the relationship information mining module, used to perform the step S400;

[0074] The prediction module is used to execute the step S500.

[0075] Furthermore, the feature representation module includes a bidirectional GRU module, which can independently learn the context information of each sequence, including the context of the subsequence. The bidirectional GRU module comprehensively learns the information of different subsequences along the sequence direction, ensuring the integrity and sequence of information during the model training process.

[0076] Among them, the combined relational information mining module includes a RetNet module, which has 24 layers connected in series. The crucial recursive representation module in the RetNet module is a variant of self-attention, which introduces the D matrix and the GN module. The exponential decay factor Y in the D matrix enables RetNet to retain the semantic influence of sequence proximity during the learning process. In addition, the GN module introduces nonlinearity to facilitate parallel training of modules. In essence, the addition of the RetNet module significantly enhances the model's understanding of the sequence and increases the training speed. Compared with the traditional Transformer position encoding, this matrix is ​​more consistent with sequence processing in terms of position perception and is more suitable for the task of predicting the binding of peptides to TCRs.

[0077] Among them, after the RetNet module outputs the relationship matrix of each pair of sequences, it enters the 3-layer linear layer, 2-layer ReLU activation, and 1 random downsampling in the prediction module in turn. Finally, the result passes through a Sigmoid activation layer to map the value to [0, 1], indicating the binding probability between the input peptide and the TCR molecule.

[0078] The above-mentioned combination relationship prediction method can be converted into software program instructions, which can be implemented by using a combination relationship prediction system including a processor and a memory, or can be implemented by computer instructions stored in a non-transitory computer-readable storage medium. The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform some steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0079] Example 1

[0080] A method for predicting the binding relationship between a peptide and a TCR molecule based on a large-scale language model comprises the following steps:

[0081] S100, obtaining a polypeptide sequence RAKFKQLL of a binding relationship to be tested and a CDR3 fragment sequence ASSFTGGTGELF on a TCR molecule, and preprocessing the sequences to be tested to obtain target input sequences RAKFXXXXXXKQLL and ASSFTGGTGELFXXXXXXXX;

[0082] S200, feature encoding each amino acid residue in the target input sequence, and outputting a coding matrix for each sequence;

[0083] S300, inputting the encoding matrices into the bidirectional GRU modules in the trained combined relationship prediction model respectively, and outputting the global feature matrix of each sequence;

[0084] S400, after merging the global feature matrices of each pair of sequences, the obtained matrix is ​​input into a large-scale language model RetNet in a trained combined relationship prediction model, and a relationship matrix of each pair of sequences is output;

[0085] S500, inputting the relationship matrix of each pair of sequences into a binding relationship prediction model, and outputting the binding probability of the polypeptide sequence of the binding relationship to be tested and the CDR3 fragment sequence on the TCR molecule as 0.8197.

[0086] Experimental verification

[0087] In order to verify the effectiveness of the prediction method of the present invention, the present invention evaluates the AUC and AUPR of several prediction tools (TEIM, TEINet, PanPep) newly released in 2023 under three different conditions in the test set. Figure 2 As shown in Figure A, the four tools were directly run on the test set, and the results showed that the method of the present invention exhibited the best prediction performance, with an AUC of 0.8164 and an AUPR of 0.8100, followed by TEINet. Secondly, in order to eliminate the bias in the test set that may appear in the training set of the method of the present invention but not in the reference tool, the present invention extracted a total of 14,551 binding pairs containing 25 peptides based on the training sets of the four tools. Figure 2 B shows that even in the crossed test set, the method of the present invention maintains its leading position, with AUC / AUPR of 0.8671 and 0.8665 respectively. This shows that the predictive advantage of the method of the present invention is still reliable on a completely fair test set. Finally, considering the good performance of the PanPep paper in its unseen set, the present invention conducted a further "fair" testing scheme. PanPep is the only one of the three reference tools that provides all positive and negative samples during training, and the data is divided into visible and invisible sets. The present invention retrained and tested the method of the present invention using the training and test data of PanPep. Figure 2 The results in C show that the method of the present invention achieves excellent prediction performance on both the visible set and the invisible set. The excellent results of the invisible set may be attributed to the use of different negative sample filling strategies by PanPep, confirming the superior predictive ability of the method of the present invention in peptide-TCR molecule binding.

[0088] In order to explore the binding characteristics of each peptide, the present invention conducted a detailed analysis of the binding preferences of the peptides based on the existing binding data. Specifically, the present invention extracted a set of peptides with more than 100 real binding CDR3 sequences from all collected binding pairs. The probability of amino acids appearing at each site was visualized using the "ggseqlogo" package. A total of 99 peptides that met the specific requirements were extracted for further analysis. After visualizing the frequency of each binding site, as shown in Figure 3 A. Although most CDR3 data are biased by insufficient length at the termini, noteworthy observations include the trend for peptides to show strong binding preferences in the head of CDR3. In contrast, a broader range of binding preferences was observed in the middle and tail of CDR3.

[0089] To evaluate the ability of tools to characterize peptide binding preferences, we extracted the top N predicted binding scores of CDR3 sequences from the test set and generated a ggseqlogo graph for each tool (N represents the number of actual binding compounds in the test set). As an example, Figure 3 B illustrates the case of TPRVTGGGAM. Figure 3 In the first row of B, the first image represents the ggseqlogo graphs of all collected TPRVTGGGAM complexes, while the second image shows the ggseqlogo graphs of all positive samples in the test set. By comparing the trends of CDR3 binding to amino acids at positions 1 / 2 / 10 / 11 / 12 / 13, it is clear that the method of the present invention shows the best reproducibility in reproducing the original binding preference of the polypeptide.

[0090] In order to further evaluate the anti-interference ability of the method of the present invention, the present invention extracted a validation data set, including 5 polypeptide sets with the largest number of binding pairs in the test set. From these 5 polypeptides, 100 non-repetitive CDR3 sequences were randomly extracted from the test set respectively, and a total of 500 CDR3 sequences were obtained. Then the method of the present invention was used to predict the binding of each CDR3 sequence to the 5 peptides. The statistical ranking of each CDR3 on the 5 polypeptides was obtained. Ideally, the original binding sequence of CDR3 should be ranked 1st. The predicted ranking of the real binders of the 5 polypeptides is as follows: Figure 4 As shown in Figure 2, all CDR3s can still recognize their original binders and rank the prediction value first among all binding pairs. Figure 4 It can be observed that many CDR3s also ranked other peptides higher. This observation is not surprising, as these are the top five peptides with the most bound compounds, and their biological significance suggests that they are highly toxic and likely to elicit responses from other CDR3s.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the embodiments of the present invention are described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations shall fall within the scope defined by the appended claims.

Claims

1. A method for predicting the binding relationship between peptides and TCR molecules based on a large-scale language model, characterized in that: The following steps are involved: S100, obtaining a polypeptide sequence of a binding relationship to be tested and a CDR3 fragment sequence on a TCR molecule, and preprocessing the sequences to be tested respectively to obtain a target input sequence; S200, feature encoding each amino acid residue in the target input sequence, and outputting a coding matrix for each sequence; S300, inputting the encoding matrices into the bidirectional GRU modules in the trained combined relationship prediction model respectively, and outputting the global feature matrix of each sequence; S400, after merging the corresponding global feature matrices of each sequence, inputting them into the trained large-scale language model RetNet in the combined relationship prediction model, and outputting the relationship matrix of each pair of sequences; S500, inputting the relationship matrix of each pair of sequences into a binding relationship prediction model, and outputting the binding probability of the polypeptide sequence of the binding relationship to be tested and the CDR3 fragment sequence on the TCR molecule.

2. The method for predicting the binding relationship between polypeptides and TCR molecules based on a large-scale language model according to claim 1, characterized in that: In step S100, the preprocessing specifically includes: filling the sequence to be tested with "X" residues to align the length of the sequence to be tested.

3. The method for predicting the binding relationship between polypeptides and TCR molecules based on a large-scale language model according to claim 1, characterized in that: Step S200 includes: According to 28 features extracted from the AAIndex database that have been successfully applied in the field of protein binding, the physical and chemical characteristics of each amino acid residue in the target input sequence are encoded.

4. The method for predicting the binding relationship between polypeptides and TCR molecules based on a large-scale language model according to claim 3, characterized in that: The 28 features that have been successfully applied in the field of protein binding are CHOP780202, CIDH920103, CIDH920105, FAUJ880109, GEIM800106, KANM800102, KLEP840101, KRIW710101, PALJ810107, QIAN880123, RACS770103, RADA880108, ZIMJ680104, AURR980120, MUNV940103, NADH010104, FAUJ880111, FINA910104, GEIM800104, NADH010106, LIFS7 90101, MEEJ800101, OOBM770102, GUYH850105, ROSM880102, SWER830101, ZIMJ680102, MIYS990104.

5. The method for predicting the binding relationship between polypeptide and TCR molecule based on a large-scale language model according to claim 1, characterized in that: Step S300 includes: Input the encoding matrix of the peptide into the bidirectional GRU module to obtain the global feature matrix of the peptide; The encoding matrix of the CDR3 fragment on the TCR molecule is input into the bidirectional GRU module to obtain the global feature matrix of CDR3.

6. The method for predicting the binding relationship between a polypeptide and a TCR molecule based on a large-scale language model according to claim 1, characterized in that: Step S400 is specifically as follows: The global feature matrices of the peptide and CDR3 segments were spliced; The concatenated matrix is ​​input into a 24-layer cascaded RetNet module, and the relationship matrix of each pair of sequences is output.

7. The method for predicting the binding relationship between a polypeptide and a TCR molecule based on a large-scale language model according to claim 1, characterized in that: Step S500 is specifically as follows: Input the relationship matrix of each pair of sequences into two layers of linear layers and ReLU layers connected in series in the combined relationship prediction model; After another linear layer and a Sigmoid layer, the value is mapped to [0, 1], indicating the binding probability between the input peptide and the TCR molecule.

8. A device for predicting the binding relationship between peptides and TCR molecules based on a large-scale language model, characterized in that: comprising at least one processor; and at least one memory in communication with the processor, wherein: The memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the method for predicting the binding relationship between a polypeptide and a TCR molecule based on a large-scale language model according to any one of claims 1 to 7, comprising the following functional modules: A data acquisition and preprocessing module, used to execute the step S100; A feature encoding module, used to execute step S200; A feature representation module, used to execute step S300; Combined with the relationship information mining module, used to perform the step S400; The prediction module is used to execute the step S500.

9. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, which enable the computer to execute the method for predicting the binding relationship between polypeptides and TCR molecules based on a large-scale language model as described in any one of claims 1 to 7.

Citation Information

Patent Citations

  • Method and system for predicting combination of T cell receptor and epitope

    CN114360644A

  • Method for establishing human source alpha beta TCR and pMHC combination probability prediction model

    CN115188410A

  • Reasoning acceleration method of large language model based on hybrid neural network structure

    CN117787410A

  • Methods and systems for assessing immune cell receptors and antigens

    WO2023114433A1

  • KR710101

Cited By

  • TCR-pMHC combined prediction method, model and system based on deep learning

    CN121215045A

  • Target spot determination method, electronic equipment, storage medium and program product

    CN121260259A

  • Target point determination method, electronic device, storage medium, and program product

    CN121260259B