Model for predicting high-affinity mutation site of antibody as well as construction method and application of model

Abstract

The invention relates to a model for predicting a high-affinity mutation site of an antibody as well as a construction method and application of the model. The construction method comprises the steps that an amino acid sequence of a mutant antibody with improved affinity is used as a training set, the training set comprises the amino acid sequence of the mutant antibody, a mutation site and amino acid information after mutation, the amino acid after mutation is used as a prediction correct target, the training set is adopted to train an ESM model, and the ESM model is obtained. The training comprises the steps of shielding the amino acid sequence of the mutation site of the mutation antibody by using a shielding language model, then predicting the amino acid sequence of the shielding site, outputting the probability of predicting a correct target, and adjusting the model according to an output result to obtain a model for predicting the high-affinity mutation site of the antibody. The model for predicting the high-affinity mutation site of the antibody is trained and optimized based on the ESM model, and can be effectively applied to an antibody affinity maturation technology to predict an effective mutation site combination.

Description

Technical Field

[0001] The invention belongs to the field of biotechnology and relates to a model for predicting antibody high-affinity mutation sites and a construction method and application thereof. Background Art

[0002] Antibodies are important proteins in the immune system that are used to recognize and neutralize foreign substances (such as pathogens, viruses, and bacteria). Antibody molecules are usually composed of four polypeptide chains, including two heavy chains and two light chains. These chains form a Y-shaped structure in the molecular structure, which enables antibodies to bind to specific antigens. The role of antibodies in immune responses is to bind to specific antigens through their variable regions, thereby neutralizing or marking these antigens for recognition and elimination by other immune cells. The diversity of antibodies is produced through genetic recombination and mutation, which enables the immune system to recognize an almost unlimited variety of antigens. Antibodies have a wide range of applications in medicine and scientific research, including disease diagnosis, vaccine development, and treatment. For example, monoclonal antibodies are widely used in cancer treatment. They can specifically target antigens on the surface of cancer cells, thereby inhibiting the growth and spread of cancer cells.

[0003] The affinity of an antibody refers to the binding force between the antibody and the antigen epitope or antigenic determinant. In the development and application of antibodies, affinity is an important indicator to measure their biological activity and clinical value. Antibodies with high affinity can significantly reduce drug dosage, reduce side effects and save costs. Affinity maturation refers to the gradual enhancement of the strength of antibody binding to the antigen during the immune response. The affinity maturation process of antibodies is achieved through high-frequency somatic gene mutations, which are mainly concentrated in the antibody CDR region. The core idea is to screen out antibody variants with higher affinity through random or directed mutations, mainly including single-point saturation mutations at each amino acid site in the CDR region, constructing a single-point saturation mutation plasmid library of the parent antibody, and obtaining single mutation points that can improve affinity. The single mutation points can be further combined to obtain a series of high-affinity mutants. For example, CN115925947A discloses an affinity maturation method and affinity maturation of an anti-human PD-L1 single-domain antibody, which obtains a high-affinity anti-human PD-L1 single-domain antibody by combining single-point saturation mutations that fully cover the CDR region with an ultra-high-throughput expression screening system of the mammalian system.

[0004] However, how to combine single-point mutations to obtain effective multi-point mutations is still one of the difficult problems in this field. Blind combination will lead to decreased affinity. Therefore, developing an effective method to predict the effect of combined mutations on antibody affinity is of great significance to the field of antibody development. Summary of the invention

[0005] In view of the deficiencies of the prior art and actual needs, the present invention provides a model for predicting the affinity of mutant antibodies and a method for constructing and applying the same, and develops a model for predicting high-affinity mutation sites of antibodies to achieve rapid and accurate prediction of high-affinity mutation site combinations of antibodies and promote the development of affinity maturation technology.

[0006] To achieve this object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for constructing a model for predicting high-affinity mutation sites of antibodies, the construction method comprising:

[0008] The amino acid sequence of the mutant antibody with improved affinity is used as a training set, wherein the training set includes the amino acid sequence, mutation site and amino acid information after mutation of the mutant antibody, and the amino acid after mutation is used as the correct prediction target; the ESM model is trained using the training set, wherein the training includes masking the amino acid sequence of the mutation site of the mutant antibody using a masked language model, and then predicting the amino acid sequence of the masked site, outputting the probability of predicting the correct target, and adjusting the model according to the output result to obtain a model for predicting the high-affinity mutation site of the antibody.

[0009] In the present invention, a model for predicting high-affinity mutation sites of antibodies is constructed based on ESM model training optimization. The resulting model can effectively predict and evaluate the naturalness of different mutation sequences, and then infer the impact of mutations on antibody affinity. It can be effectively applied to antibody affinity maturation technology to predict effective mutation site combinations.

[0010] Preferably, when there are more than one masked mutation sites, the average of the sum of probabilities of multiple predicted correct targets is taken as the output result.

[0011] Preferably, the adjustment is performed until the output result is not lower than a preset threshold, and the threshold is 0.5.

[0012] In a second aspect, the present invention provides a model for predicting antibody high-affinity mutation sites, wherein the model is constructed by the method for constructing a model for predicting antibody high-affinity mutation sites described in the first aspect.

[0013] In a third aspect, the present invention provides a method for predicting a combination of high-affinity mutation sites of an antibody, the method comprising:

[0014] The amino acid sequence of the mutant antibody is input into the model for predicting antibody high-affinity mutation sites described in the second aspect, all single mutation sites are masked using a masked language model, and then the amino acid sequence of the masked site is predicted, and the probability of predicting the correct target is output, and the prediction results of multiple mutation sites are added and averaged as the score of the mutation site combination, and the mutation site combination with a high score is selected as the candidate antibody high-affinity mutation site combination.

[0015] In the present invention, a method for predicting combinations of high-affinity mutation sites of antibodies is further developed. Based on the constructed prediction model, the effects of different mutation combinations on antibody affinity can be quickly and accurately predicted, thereby guiding the construction of high-affinity antibodies.

[0016] In the present invention, the mutation site combinations can be ranked from high to low based on the scores, and then the mutation site combinations with high scores are selected for affinity testing and verification to guide the construction of high-affinity antibodies.

[0017] Preferably, the method further comprises performing single-point saturation mutation on the predicted antibody and testing affinity to obtain mutant antibodies and single-point mutation information.

[0018] Preferably, the method further comprises the step of performing affinity testing on an antibody containing a combination of high-affinity mutation sites of the candidate antibody.

[0019] It is understood that the present invention can be used to perform predictions for any antibody.

[0020] Preferably, the antibody to be predicted includes CD138 antibody.

[0021] Preferably, the amino acid sequence of the heavy chain variable region of the antibody to be predicted includes the sequence shown in SEQ ID NO.1.

[0022] Preferably, the amino acid sequence of the light chain variable region of the antibody to be predicted includes the sequence shown in SEQ ID NO.2.

[0023] Preferably, the mutation site combination includes a heavy chain mutation site combination and a light chain mutation site combination, the heavy chain mutation site combination includes D54F and S57F, and the light chain mutation site combination includes a combination of L46G, Y50G, I53F and N92Y, or a combination of L46G, Y50G, I53W and N92Y.

[0024] In a fourth aspect, the present invention provides a high-affinity antibody, wherein the antibody contains a combination of mutation sites predicted by the method for predicting a combination of mutation sites of high-affinity antibodies described in the third aspect.

[0025] Specifically, the present invention provides a high-affinity CD138 antibody, wherein the high-affinity CD138 antibody is mutated based on the heavy chain variable region with an amino acid sequence as shown in SEQ ID NO.1 and the light chain variable region with an amino acid sequence as shown in SEQ ID NO.2 according to the mutation site combination predicted by the method for predicting antibody high-affinity mutation site combination described in the third aspect.

[0026] Preferably, the mutations in the heavy chain variable region include D54F and S57F.

[0027] Preferably, the mutations in the light chain variable region include a combination of L46G, Y50G, I53F and N92Y, or a combination of L46G, Y50G, I53W and N92Y.

[0028] In a fifth aspect, the present invention provides a device for predicting antibody high-affinity mutation site combinations, the device comprising a single-point mutation testing unit, a mutation combination prediction unit, and a verification unit;

[0029] The single point mutation test unit is used to perform:

[0030] Perform single-point saturation mutation on the predicted antibody and test affinity to obtain mutant antibody and single-point mutation information;

[0031] The mutation combination prediction unit is used to perform the following steps:

[0032] Inputting the amino acid sequence of the mutant antibody into the model for predicting antibody high-affinity mutation sites described in the second aspect, outputting the probability of predicting the correct target, adding and averaging the prediction results of multiple mutation sites as the score of the mutation site combination, and selecting the mutation site combination with a high score as the candidate antibody high-affinity mutation site combination;

[0033] The verification unit is used to perform the following steps:

[0034] Affinity testing is performed on antibodies containing a combination of high-affinity mutation sites of the candidate antibodies.

[0035] In a sixth aspect, the present invention provides an electronic device, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for predicting antibody high-affinity mutation site combinations described in the third aspect.

[0036] In a seventh aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the program is executed by a processor, the method for predicting antibody high-affinity mutation site combinations described in the third aspect is implemented.

[0037] Compared with the prior art, the present invention has at least the following beneficial effects:

[0038] In the present invention, a model for predicting antibody high-affinity mutation sites is constructed based on ESM model training optimization. The resulting model can effectively predict and evaluate the naturalness of different mutation sequences, and then infer the impact of mutations on antibody affinity. It can be effectively applied to antibody affinity maturation technology to predict effective mutation site combinations and guide the construction of high-affinity antibodies. DETAILED DESCRIPTION

[0039] The technical solution of the present invention is further described below by specific implementation methods. However, the following examples are only simplified examples of the present invention and do not represent or limit the scope of protection of the present invention. The scope of protection of the present invention shall be subject to the claims.

[0040] If no specific techniques or conditions are specified in the examples, the techniques or conditions described in the literature in the field or the product instructions are used. If no manufacturer is specified for the reagents or instruments used, they are all conventional products that can be purchased through regular channels.

[0041] Transformer is a deep learning model based on the self-attention mechanism. The Transformer model mainly consists of two parts: the encoder and the decoder. The encoder is responsible for encoding the input sequence into a vector of fixed length, and the decoder predicts the next output based on the encoder's output and the generated partial output sequence. In the encoder, the input sequence is first converted into a vector representation through the word embedding layer. Then, these vectors pass through multiple self-attention layers and feedforward neural network layers. The self-attention layer is used to calculate the degree of association between each position in the sequence, and the feedforward neural network layer performs further nonlinear transformation on the representation of each position. The decoder also contains multiple self-attention layers and feedforward neural network layers, but unlike the encoder, the self-attention layer in the decoder uses a mask mechanism to prevent future information from being seen when predicting the current position. In addition, the self-attention layer in the decoder also takes into account the encoder's output, realizing information interaction between the encoder and the decoder. The Transformer model uses a multi-head attention mechanism during training, which divides the input sequence into multiple heads and performs self-attention calculations separately.

[0042] In the field of amino acid sequence analysis, the Masked Language Model (MLM) is a pre-training strategy that aims to improve the model's ability to understand protein sequences through masking and prediction techniques. Specifically, the model masks some amino acids in the original amino acid sequence and then uses contextual information to restore the masked amino acids to learn the intrinsic structure and biological semantics of the sequence. In the implementation process, MLM first randomly selects several positions in the amino acid sequence and replaces the amino acids at these positions with specific masking marks. Subsequently, the model is trained to recognize and predict the original amino acids at these masked positions. This process not only enables the model to capture the sequence dependencies between amino acids, but also enables it to gain insight into the role of amino acids in protein function and structure formation. In this way, the MLM pre-trained model can gain an understanding of the deep-level features of the amino acid sequence, which is crucial for a variety of bioinformatics tasks. For example, in the fields of protein function prediction, structure prediction, sequence variation analysis, and protein-protein interaction prediction, the MLM pre-trained model has shown better performance.

[0043] The ESM model (Evolutionary Scale Modeling) is a Transformer-based protein language model, which is mainly used to predict and generate the structure and function of proteins. The ESM model uses deep learning technology to treat protein sequences as a language, each amino acid as a character, and uses autoregressive neural networks to learn the statistical laws of this language. Among them, ESM-2 (Evolutionary Scale Modeling2) is a new MLM language model developed by Facebook AI Research, which aims to understand and predict the structure and function of protein sequences through pre-training and fine-tuning. It is based on the Transformer architecture and uses an 8-layer Transformer encoder. The model scale reaches 15 billion parameters. ESM-2 performs well in predicting the three-dimensional structure of proteins and can make predictions at atomic resolution, providing a powerful tool for bioinformatics and protein engineering.

[0044] In the present invention, a model for predicting antibody high-affinity mutation sites is constructed based on ESM model training optimization. The resulting model can effectively predict and evaluate the naturalness of different mutation sequences, and then infer the impact of mutations on antibody affinity. It can be effectively applied to antibody affinity maturation technology to predict effective mutation site combinations and guide the construction of high-affinity antibodies.

[0045] In one embodiment of the present invention, a device for predicting antibody high-affinity mutation site combinations can be provided, characterized in that the device includes a single-point mutation testing unit, a mutation combination prediction unit and a verification unit; the single-point mutation testing unit is used to perform the following steps: performing single-point saturation mutation on the antibody to be predicted, and testing affinity to obtain mutant antibodies and single-point mutation information; the mutation combination prediction unit is used to perform the following steps: inputting the amino acid sequence of the mutant antibody into a model for predicting antibody high-affinity mutation sites, outputting the probability of predicting the correct target, adding and averaging the prediction results of multiple mutation sites as the score of the mutation site combination, and selecting a mutation site combination with a high score as a candidate antibody high-affinity mutation site combination; the verification unit is used to perform the following steps: performing affinity testing on antibodies containing the candidate antibody high-affinity mutation site combination.

[0046] In another specific embodiment of the present invention, an electronic device may be provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method for predicting antibody high-affinity mutation site combinations.

[0047] In another specific embodiment of the present invention, a computer-readable storage medium may be provided, on which a computer program is stored, and when the program is executed by a processor, the method for predicting antibody high-affinity mutation site combinations is implemented.

[0048] Example 1

[0049] This example uses CD138 antibody as an example to predict mutation site combinations.

[0050] The heavy chain amino acid sequence of CD138 antibody is shown in SEQ ID NO.1, and the light chain amino acid sequence is shown in SEQ ID NO.2. The monosaturated mutation sites of the heavy chain include H_G50W (representing the mutation of G to W at position 50 of the heavy chain, and the remaining mutations can be deduced by analogy), H_I52W, H_D54F, H_D54Y, H_S57W and H_S57F, and the light chain mutation sites include L_L46G, L_I48F, L_I48G, L_Y49W, L_Y49F, L_Y50G, L_Y50W, L_Y50G, L_S52G, L_S52W, L_I53W, L_I53F, L_N92F, L_N92Y and L_P95G.

[0051] Based on the ESM model, model training optimization and prediction of mutation site combinations were performed. Taking G50W as an example for model training, the original sequence replaced site 50 in the heavy chain (H) with a mask. After prediction by the esm model, the target of this site was W (mask(H_G50), W).

[0052] Specifically, the mutant antibody sequence is input into seq_mask(H_G50).

[0053] (1) Embedding layer:

[0054] emb=embedding(seq_mask)+position_embedding(seq_mask_index), get emb.

[0055] (2) Encoder:

[0056] Then it passes through the self-attention layer and the feedforward network layer. Then, the composite state tensor EmbeddingVec1 is multiplied by three weight matrices to obtain the three matrices Q, K, and V as the input of the multi-head attention layer. Since the multi-head design allows each state to be correlated with the state at other times, there are multiple such Q, K, and V. The calculation formula of the i-th one is as follows:

[0057] Q = emb.Wq;

[0058] K = emb.Wk;

[0059] V = emb.Wv.

[0060] The attention tensor of the i-th output of the multi-head attention layer includes the following calculation formula:

[0061]

[0062] Among them, dK is the dimension of K.

[0063] Then, the attention tensors are concatenated and mapped to the original input space as the output MultiHead of the multi-head attention layer.

[0064] MultiHead=concat(head1,head2....headh)W;

[0065] AttentionOut=LayerNorm(EmbeddingVec1+MutiHead1).

[0066] The obtained multi-head attention layer output AttentionOut1 is then input into the feedforward network layer to obtain FNNOut1. The feedforward network layer is composed of a neural network nested with a ReLU function and then a neural network nested. The calculation formula is as follows:

[0067] FFNOut=max(0,AttentionOut*W1+b1)*W2+b2.

[0068] The feedforward network layer also contains a residual connection structure. After the residual connection, the final output EncoderVec of the encoder is obtained. The calculation formula of the encoding tensor EncoderVec is as follows:

[0069] EncoderVec=LayerNorm(AttentionOut1+FFNOut1).

[0070] (3) Conversion layer:

[0071] The output is mapped into character probabilities. During training, the model is made to learn high-scoring probabilities such as H_G50W, and the model output is made to be W as much as possible. During prediction, if the combination of H_G50W and H_I52W is scored, the output is the average of the probability of outputting W at the 50th site of the H chain and the probability of outputting W at the 52nd site of the H chain.

[0072] The combinations with relatively high scores based on the light chain are: L46G, Y50G, I53F and N92Y, with a score of 0.5152; or, L46G, Y50G, I53W and N92Y, with a score of 0.5443; the combinations with relatively high scores based on the heavy chain are: D54F and S57F, with a score of 0.7.

[0073] Example 2

[0074] This example tests the effect of the mutation combination obtained in Example 1 on affinity.

[0075] (1) Preparation of antibodies with specific mutation combinations

[0076] Two CD138 antibodies can be designed according to the mutation combination. Based on the heavy and light chains of SEQ ID NO.1 and SEQ ID NO.2, mutant antibody 1 is a combination of heavy chain D54F and S57F mutations, and light chain L46G, Y50G, I53F and N92Y mutations; mutant antibody 2 is a combination of heavy chain D54F and S57F mutations, and light chain L46G, Y50G, I53W and N92Y mutations.

[0077] 1) Primer design: Design and synthesize primers for antibody light and heavy chain genes, and add appropriate restriction endonuclease recognition sequences to the 5' end of the primers.

[0078] 2) Gene synthesis: Use the designed primers to splice the genes by PCR reaction to obtain the light and heavy chain DNA sequences of the antibody.

[0079] 3) Enzyme digestion reaction: The PCR product and the cloning vector are digested with the same restriction endonuclease to produce sticky ends.

[0080] 4) Ligation reaction: connect the antibody gene fragment after enzyme digestion to the vector using DNA ligase.

[0081] 5) Transformation: Transform the ligation product into competent E. coli.

[0082] 6) Screening and identification: Positive clones are screened by antibiotics, and colony PCR and sequence analysis are performed to confirm the correctness of the inserted fragment.

[0083] 7) Expression and purification: The verified correct plasmid is transformed into a eukaryotic or prokaryotic expression system to induce the expression of antibody protein, and the protein is purified by affinity chromatography and other methods.

[0084] (2) Affinity test

[0085] 1) Ligand capture: The purified antibody was diluted to an appropriate concentration with a running reagent and immobilized on the surface of the PAHC200M chip.

[0086] 2) Analyte dilution: The analyte is diluted 3 times with the running reagent. The diluted analytes are injected onto the chip surface from the lowest concentration to the highest concentration, and the binding and dissociation times are corresponding.

[0087] 3) Baseline balance: Before running the analyte, the solution is circulated several times to balance the baseline. The binding and dissociation steps are all performed in the running reagent.

[0088] 4) Chip regeneration: After all analyte concentration cycles are completed, regenerate with 10 mM Glycine, pH 2.0 for 3 times, 20 s each time, to wash away the ligand and undissociated analytes.

[0089] 5) Data collection and analysis: After collecting the data, use Carterra software for kinetic analysis. Use a 1:1 Langmuir binding model or other appropriate models to fit the data. Calculate the KD value.

[0090] The affinity results are shown in Table 1, indicating that the combination of mutation sites predicted by the present invention can significantly improve the affinity of the antibody.

[0091] Table 1

[0092]

[0093] In summary, in the present invention, a model for predicting antibody high-affinity mutation sites is constructed based on ESM model training optimization. The resulting model can effectively predict and evaluate the naturalness of different mutation sequences, and then infer the impact of mutations on antibody affinity. It can be effectively applied to antibody affinity maturation technology to predict effective mutation site combinations and guide the construction of high-affinity antibodies.

[0094] The applicant declares that the above is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention shall fall within the protection scope and disclosure scope of the present invention.

Claims

1. A method for constructing a model for predicting high-affinity mutation sites of antibodies, characterized in that: The construction method comprises: The amino acid sequence of the mutant antibody with improved affinity is used as a training set, wherein the training set includes the amino acid sequence, mutation site and amino acid information after mutation of the mutant antibody, and the amino acid after mutation is used as the correct target for prediction; The ESM model is trained using a training set. The training includes using a masked language model to mask the amino acid sequence of the mutation site of the mutant antibody, then predicting the amino acid sequence of the masked site, outputting the probability of predicting the correct target, and adjusting the model according to the output result to obtain a model for predicting antibody high-affinity mutation sites.

2. The method for constructing a model for predicting high-affinity mutation sites of antibodies according to claim 1, characterized in that: When there are more than one masked mutation sites, the average of the sum of probabilities of multiple predicted correct targets is used as the output result; Preferably, the adjustment is performed until the output result is not lower than a preset threshold, and the threshold is 0.

5.

3. A model for predicting high-affinity mutation sites of antibodies, characterized in that: The model is constructed by the method for constructing a model for predicting antibody high-affinity mutation sites according to claim 1 or 2.

4. A method for predicting antibody high-affinity mutation site combinations, characterized in that: The method comprises: The amino acid sequence of the mutant antibody is input into the model for predicting antibody high-affinity mutation sites as described in claim 3, all single mutation sites are masked using a masked language model, and then the amino acid sequence of the masked site is predicted, and the probability of predicting the correct target is output, and the prediction results of multiple mutation sites are added and averaged as the score of the mutation site combination, and the mutation site combination with a high score is selected as the candidate antibody high-affinity mutation site combination.

5. The method for predicting antibody high-affinity mutation site combinations according to claim 4, characterized in that: The method further comprises performing single-point saturation mutation on the predicted antibody and testing affinity to obtain mutant antibodies and single-point mutation information.

6. A device for predicting antibody high-affinity mutation site combinations, characterized in that: The device comprises a single-point mutation testing unit, a mutation combination prediction unit and a verification unit; The single point mutation test unit is used to perform: Perform single-point saturation mutation on the predicted antibody and test affinity to obtain mutant antibody and single-point mutation information; The mutation combination prediction unit is used to perform the following steps: Inputting the amino acid sequence of the mutant antibody into the model for predicting antibody high-affinity mutation sites according to claim 4, outputting the probability of predicting the correct target, adding and averaging the prediction results of multiple mutation sites as the score of the mutation site combination, and selecting the mutation site combination with a high score as the candidate antibody high-affinity mutation site combination; The verification unit is used to perform the following steps: Affinity testing is performed on antibodies containing a combination of high-affinity mutation sites of the candidate antibodies.

7. A high-affinity antibody, characterized in that The antibody contains the combination of mutation sites predicted by the method for predicting a combination of mutation sites for high affinity antibodies according to any one of claims 4 to 5.

8. A high-affinity CD138 antibody, characterized in that: The high-affinity CD138 antibody is mutated by a combination of mutation sites predicted by the method for predicting a combination of high-affinity mutation sites for antibodies according to any one of claims 4 to 5 based on the amino acid sequence of the heavy chain variable region shown in SEQ ID NO.1 and the amino acid sequence of the light chain variable region shown in SEQ ID NO.2; Preferably, the mutations in the heavy chain variable region include D54F and S57F; Preferably, the mutations in the light chain variable region include a combination of L46G, Y50G, I53F and N92Y, or a combination of L46G, Y50G, I53W and N92Y.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for predicting antibody high-affinity mutation site combinations as described in any one of claims 4-5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for predicting a combination of high-affinity mutation sites of an antibody as described in any one of claims 4 to 5 is implemented.

Images