Method, apparatus and electronic device for obtaining conformational antigen-antibody complex

Abstract

The present disclosure provides a method, apparatus, and electronic device for obtaining the conformation of an antigen-antibody complex, relating to the field of artificial intelligence technology, and specifically to the field of bio-computation technology. The specific implementation scheme is as follows: obtaining an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence; performing a multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain; and obtaining the conformation of the antigen-antibody complex based on the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, specifically to the field of biocomputing technology, and particularly to a method, apparatus, and electronic device for obtaining the conformation of an antigen-antibody complex. Background Art

[0002] Protein conformation, especially the conformation of protein complexes, is of great significance in life science research. It not only reveals the mechanism of protein-protein interactions but also provides valuable tools and insights for fields such as drug design, disease research, and genomics.

[0003] The conformation of the antigen-antibody complex is a special type of protein complex. The conformation of the antigen-antibody complex determines its function in the human body and also determines antibody activity, which plays an important role in antibody drug design. However, the conformations of the vast majority of antibody-antigen complexes are unknown, and the cost of obtaining the conformation of the antigen-antibody complex based on experimental methods is relatively high. Summary of the Invention

[0004] The present disclosure provides a method, apparatus, and electronic device for obtaining the conformation of an antigen-antibody complex.

[0005] According to one aspect of the present disclosure, a method for obtaining the conformation of an antigen-antibody complex is provided, including: obtaining an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence; performing multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain; and obtaining the conformation of the antigen-antibody complex based on the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation.

[0006] According to another aspect of the present disclosure, an apparatus for obtaining the conformation of an antigen-antibody complex is provided, including: a first acquisition module for obtaining an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence; an MSA module for performing multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain; and a second acquisition module for obtaining the conformation of the antigen-antibody complex based on the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation.

[0007] According to another aspect of the present disclosure, an electronic device is provided, including: at least one processor; and a memory communicatively connected to the at least one processor; where the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the method for obtaining the conformation of an antigen-antibody complex according to the embodiment of the above-mentioned one aspect.

[0008] According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, on which a computer program / instruction is stored, and the computer instructions are used to cause the computer to execute the method for obtaining the complex conformation of antigen-antibody described in the embodiment of the above aspect.

[0009] According to another aspect of the present disclosure, there is provided a computer program product including a computer program / instruction, and when the computer program / instruction is executed by a processor, the method for obtaining the complex conformation of antigen-antibody described in the embodiment of the above aspect is implemented.

[0010] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0012] Figure 1 is a schematic flowchart of a method for obtaining the complex conformation of antigen-antibody provided by an embodiment of the present disclosure;

[0013] Figure 2 is a schematic flowchart of another method for obtaining the complex conformation of antigen-antibody provided by an embodiment of the present disclosure;

[0014] Figure 3 is a schematic flowchart of another method for obtaining the complex conformation of antigen-antibody provided by an embodiment of the present disclosure;

[0015] Figure 4 is a schematic flowchart of another method for obtaining the complex conformation of antigen-antibody provided by an embodiment of the present disclosure;

[0016] Figure 5 is a schematic flowchart of optimizing the antibody sequence provided by an embodiment of the present disclosure;

[0017] Figure 6 is a schematic flowchart of another method for obtaining the complex conformation of antigen-antibody provided by an embodiment of the present disclosure;

[0018] Figure 7 is a schematic flowchart of obtaining the complex conformation provided by an embodiment of the present disclosure;

[0019] Figure 8 is a schematic structural diagram of obtaining the complex conformation provided by an embodiment of the present disclosure;

[0020] Figure 9Schematic structural diagram of an apparatus for obtaining the conformation of an antigen-antibody complex provided by an embodiment of the present disclosure;

[0021] Figure 10 Block diagram of an electronic device for implementing a method for obtaining the conformation of an antigen-antibody complex according to an embodiment of the present disclosure. Detailed implementation manners

[0022] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for clarity and conciseness, descriptions of well-known functions and structures are omitted in the following.

[0023] The following describes a method, an apparatus, and an electronic device for obtaining the conformation of an antigen-antibody complex according to an embodiment of the present disclosure with reference to the accompanying drawings.

[0024] Artificial Intelligence (AI) is a discipline that studies how to make a computer simulate certain thinking processes and intelligent behaviors of humans (such as learning, reasoning, thinking, planning, etc.). It includes both hardware-level technologies and software-level technologies. Artificial intelligence hardware technologies generally include several aspects such as computer vision technology, speech recognition technology, natural language processing technology, machine learning / deep learning, big data processing technology, and knowledge graph technology.

[0025] Bio-computing is a field that draws on the principles and mechanisms of biological systems to solve computational problems. It applies some characteristics and processes of biology to computational systems to improve computational efficiency and performance. The goal of bio-computing is to obtain inspiration from biological systems and transform it into new computational methods and technologies to solve complex problems. It has a wide range of applications in fields such as optimization, pattern recognition, data analysis, and simulation, and is constantly developing and expanding.

[0026] The method for obtaining the conformation of an antigen-antibody complex provided by the embodiments of the present disclosure can be applied in drug design, such as antibody drug design, Chimeric Antigen Receptor T-Cell Immunotherapy (CAR-T), Chimeric Antigen Receptor Nature killer cell Immunotherapy (CAR-NK), Antibody-Drug Conjugates (ADC) design, etc., as well as in the research of the mechanism of immune diseases and other fields.

[0027] Figure 1 This is a schematic flowchart of a method for obtaining the conformation of an antigen-antibody complex provided by an embodiment of the present disclosure.

[0028] As Figure 1 shown, the method for obtaining the conformation of the antigen-antibody complex may include:

[0029] S101. Obtain an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence.

[0030] It should be noted that the execution subject of the method for obtaining the conformation of the antigen-antibody complex in the embodiments of the present disclosure may be a hardware device with data processing capabilities and / or the necessary software for driving the hardware device to work. Optionally, the execution subject may include a server, a user terminal, and other intelligent devices. Optionally, the user terminal includes, but is not limited to, a mobile phone, a computer, an intelligent voice interaction device, etc. Optionally, the server includes, but is not limited to, a web server, an application server, and may also be a server of a distributed system, or a server combined with a blockchain, etc. The embodiments of the present disclosure do not make specific limitations.

[0031] It can be understood that the amino acid sequence refers to the sequence composed of a series of amino acids in a protein molecule, which is the basis of the protein structure and function. The antigen sequence refers to the sequence of a biomolecule (such as a protein) or a chemical substance that can be recognized by the immune system and trigger an immune response. For example, in immunology, the antigen sequence usually refers to the amino acid sequence of a protein antigen.

[0032] The antibody sequence is the sequence encoded by the immunoglobulin gene in the genome. Each antibody has its unique amino acid sequence and can bind to a specific antigen.

[0033] In some implementations, the amino acid sequence can be obtained based on sequencing technology and genomics methods, and / or the amino acid sequence can be obtained from a protein database. The obtained amino acid sequence includes an antigen sequence and a reference antibody sequence.

[0034] S102. Perform a multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain.

[0035] It can be understood that multiple sequence alignment (MSA) can be used to analyze the similarities and differences between different protein chains in the amino acid sequence. Through MSA, the comparison sequence of each protein chain can be obtained.

[0036] In some implementations, the amino acid sequence can be subjected to MSA based on the MSA algorithm or by using the MSA search tool, to obtain the MSA sequences similar to each protein chain in the amino acid sequence, as well as the corresponding reference conformations of the protein chains.

[0037] S103, obtain the complex conformation of the antigen-antibody according to the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation.

[0038] In some implementations, by performing feature extraction on the amino acid sequence and the MSA sequence of the protein chain, the features of the amino acid and the MSA sequence can be obtained and the features can be encoded to obtain the encoded representation of the amino acid pair and the encoded representation of the MSA sequence.

[0039] Furthermore, optimize and update the encoded representation of the amino acid pair and the encoded representation of the MSA sequence based on the reference conformation. Optionally, by obtaining the structural features of the reference conformation and fusing the structural features with the encoded representation of the amino acid pair and the encoded representation of the MSA sequence, the encoded representation of the amino acid pair and the encoded representation of the MSA sequence can be optimized and updated.

[0040] Furthermore, based on the encoded representation of the optimized MSA sequence, determine the encoded representation of the amino acid, and according to the encoded representation of the amino acid and the encoded representation of the optimized amino acid pair, predict the residue coordinates of the protein, and then construct the complex conformation of the antigen-antibody based on the residue coordinates.

[0041] It can be understood that in the protein structure, each amino acid residue has its specific three-dimensional spatial coordinates, called residue coordinates. The residue coordinates describe the position and orientation of the amino acid residue in three-dimensional space. Among them, the residue refers to the basic unit composed of amino acids.

[0042] According to the method for obtaining the complex conformation of the antigen-antibody provided by the embodiments of the present disclosure, by obtaining the amino acid sequence and subjecting the amino acid sequence to MSA, the MSA sequences of each protein chain and the corresponding reference conformations of the protein chains are obtained. Furthermore, by encoding and optimizing the amino acid sequence, the MSA sequence, and the reference conformation, the residue coordinates of the protein can be predicted, and the complex conformation can be constructed, improving the accuracy of predicting the complex conformation of the antigen-antibody and reducing the prediction cost of the complex conformation. At the same time, it also provides a basis for drug design related to antigen-antibody, helping to promote the development of antibody drug design.

[0043] Figure 2 It is a schematic flowchart of a method for obtaining the complex conformation of an antigen-antibody provided by the embodiments of the present disclosure.

[0044] As Figure 2 shown, the method for obtaining the complex conformation of the antigen-antibody may include:

[0045] S201, obtain the amino acid sequence, where the amino acid sequence includes the antigen sequence and the reference antibody sequence.

[0046] S202, perform multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain.

[0047] For the relevant content of steps S201 - S202, reference can be made to the above embodiments and will not be elaborated here.

[0048] S203, obtain the first amino acid pair encoding between amino acid pairs in the amino acid sequence.

[0049] In some implementations, based on the positional relationship of amino acids, the amino acid pairs can be encoded to obtain the first amino acid pair encoding between amino acid pairs, which is convenient for understanding the interaction and connection between amino acids, and further deepens the understanding of the protein structure.

[0050] In some implementations, the amino acid sequence can be converted into a coded representation, and then based on the position of the amino acid in the amino acid sequence, the first amino acid pair encoding can be obtained. By performing embedding layer encoding on the amino acid sequence, the embedding encoding of the amino acid sequence is obtained, and this embedding encoding is the coded representation of the amino acid sequence.

[0051] Furthermore, based on the index value of the amino acid, the amino acid can be associated with its position in the sequence. By encoding the index value of the amino acid in the amino acid sequence, the position encoding is obtained, and the embedding encoding and the position encoding are fused to obtain the first amino acid pair encoding.

[0052] S204, obtain the first MSA encoding according to the characteristics of the MSA sequence and the characteristics of the amino acid sequence.

[0053] In some implementations, by characterizing the MSA sequence and the amino acid sequence, the characteristics of the MSA sequence and the characteristics of the amino acid sequence are obtained, and the characteristics of the MSA sequence and the characteristics of the amino acid sequence are combined and encoded, and the first MSA encoding can be obtained to realize the fusion of the information of the MSA sequence and the amino acid sequence.

[0054] S205, obtain the complex conformation of the antigen - antibody according to the first amino acid pair encoding, the first MSA encoding and the reference conformation.

[0055] In some implementations, based on the reference conformation, the first amino acid pair encoding and the first MSA encoding can be optimized to obtain the optimized second amino acid pair encoding and second MSA encoding. Then, the residue coordinates are predicted for the second amino acid pair encoding and the second MSA encoding to construct the complex conformation of the antigen - antibody.

[0056] Optionally, the first amino acid pair encoding and the first MSA encoding can be optimized based on the characteristics of the reference conformation. The characteristics of the reference conformation can be encoded and spliced and fused with the first amino acid pair encoding and the first MSA encoding to obtain the second amino acid pair encoding and the second MSA encoding.

[0057] According to the method for obtaining the complex conformation of an antigen-antibody provided by the embodiments of the present disclosure, by obtaining an amino acid sequence and performing MSA on the amino acid sequence, the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain are obtained. By performing embedding layer encoding and index value encoding on the amino acid sequence, the first amino acid pair encoding is obtained. By obtaining the characteristics of the MSA sequence and the characteristics of the amino acid sequence and performing encoding, the first MSA encoding is obtained. Further, based on the reference conformation, the first amino acid pair encoding and the first MSA encoding are optimized, and the residue coordinates of the protein are predicted to construct the complex conformation, which improves the accuracy of predicting the antigen-antibody complex conformation and reduces the prediction cost of the complex conformation. At the same time, it also provides a basis for drug design related to antigen-antibodies and helps to promote the development of antibody drug design.

[0058] Figure 3 It is a schematic flowchart of a method for obtaining the complex conformation of an antigen-antibody provided by the embodiments of the present disclosure.

[0059] As Figure 3 shown, the method for obtaining the complex conformation of the antigen-antibody may include:

[0060] S301, obtain an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence.

[0061] S302, perform multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain.

[0062] S303, obtain the first amino acid pair encoding between amino acid pairs in the amino acid sequence.

[0063] S304, obtain the first MSA encoding according to the characteristics of the MSA sequence and the characteristics of the amino acid sequence.

[0064] The relevant content of steps S301 - S304 can be referred to the above embodiments and will not be elaborated here.

[0065] S305, obtain the pairing characteristics of the reference conformation, encode the pairing characteristics to obtain the second encoding of the reference conformation, and splice the first amino acid pair encoding and the second encoding to obtain the second amino acid pair encoding.

[0066] S306. Obtain the angular features of the reference conformation, encode the angular features to obtain the first encoding of the reference conformation, and concatenate the first MSA encoding and the first encoding to obtain the second MSA encoding.

[0067] In some implementations, the features of the reference conformation include, but are not limited to: pairing features and angular features. Among them, the pairing features describe the interaction features between amino acids, and the angular features reflect the spatial arrangement between amino acids.

[0068] In some implementations, by performing feature extraction on the reference conformation, the pairing features and angular features of the reference conformation can be obtained. Then, the pairing features and angular features are encoded respectively to obtain the second encoding of the pairing features and the first encoding of the angular features.

[0069] Furthermore, by concatenating the first amino acid pair encoding and the second encoding, the second amino acid pair encoding can be obtained, and the pairing features can be fused onto the second amino acid pair encoding to optimize the first amino acid pair encoding. By concatenating the first MSA encoding and the first encoding, the second MSA encoding can be obtained, and the angular features can be fused onto the second MSA encoding to optimize the first MSA encoding.

[0070] Optionally, the second encoding can be concatenated behind the first amino acid pair encoding to obtain the second amino acid pair encoding; the first encoding can be concatenated behind the first MSA encoding to obtain the second MSA encoding.

[0071] S307. Obtain the complex conformation of the antigen-antibody according to the second amino acid pair encoding and the second MSA encoding.

[0072] In some implementations, in order to perform information interaction between the second amino acid pair encoding and the second MSA encoding and further optimize the second amino acid pair encoding and the second MSA encoding, the second amino acid pair encoding and the second MSA encoding can be input into the model backbone Evoformer network. Through the Evoformer network, information interaction is performed on the second amino acid pair encoding and the second MSA encoding to obtain the optimized third amino acid pair encoding and the third MSA encoding.

[0073] It can be understood that the Evoformer network is composed of Triangle self-attention, and Triangle self-attention includes Row-wise gated attention and Column-wise gated attention.

[0074] Optionally, by inputting the second amino acid pair encoding and the second MSA encoding into the Row-wise gated attention and Column-wise gated attention, the interaction of row and column information in the second amino acid pair encoding and the second MSA encoding is realized, and the third amino acid pair encoding and the third MSA encoding are obtained.

[0075] Furthermore, based on the third amino acid pair encoding and the third MSA encoding, the complex conformation of the antigen-antibody can be obtained. By obtaining the three-dimensional atomic coordinates of the protein, that is, the residue coordinates, the spatial structure of the complex conformation can be visually presented, improving the accuracy of constructing the complex conformation of the antigen-antibody. Optionally, the three-dimensional atomic coordinates can be predicted based on the amino acid encoding, amino acid pair encoding, and structural information.

[0076] In some implementations, the amino acid encoding is the first row in the MSA encoding. By obtaining the elements of the first row of the third MSA encoding and transforming the elements of the first row, the amino acid encoding is obtained. Optionally, the elements of the first row can be transformed based on a neural network.

[0077] Furthermore, based on the amino acid encoding, the third amino acid pair encoding, and the initialized structural information, coordinate prediction can be performed to obtain the three-dimensional atomic coordinates of the protein. Optionally, the structural information can update the amino acid encoding and the third amino acid pair encoding, realizing the addition of spatial structure information in the amino acid encoding and the third amino acid pair, which helps to improve the accuracy of predicting the complex conformation.

[0078] Optionally, the amino acid encoding and the third amino acid pair encoding can be input into an Invariant Point Attention (IPA) network for processing to obtain the candidate amino acid encoding and the candidate amino acid pair encoding. Among them, the IPA network is used to construct the interaction between amino acids.

[0079] At the same time, the amino acid encoding and the structural information are input into a Chain-level Information Interaction Network (CIN) network for processing, and the candidate amino acid encoding and the candidate amino acid pair encoding are updated based on the output of the CIN network to obtain the target amino acid encoding and the target amino acid pair encoding.

[0080] Furthermore, based on the target amino acid encoding and the target amino acid pair encoding, the three-dimensional atomic coordinates are obtained, and based on the three-dimensional atomic coordinates, the complex conformation of the antigen-antibody is determined. Optionally, based on molecular docking software, using the three-dimensional atomic coordinates, the complex conformation of the antigen-antibody can be generated.

[0081] According to the method for obtaining the complex conformation of antigen and antibody provided by the embodiments of the present disclosure, by obtaining the amino acid sequence and performing MSA on the amino acid sequence, the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain are obtained. By performing embedding layer encoding and index value encoding on the amino acid sequence, the first amino acid pair encoding is obtained. By obtaining the features of the MSA sequence and the features of the amino acid sequence and performing encoding, the first MSA encoding is obtained. Further, the features of the reference conformation are used to optimize the first amino acid pair encoding and the first MSA encoding to obtain the second amino acid pair encoding and the second MSA encoding, and the Evoformer network is used to update the second amino acid pair encoding and the second MSA encoding to obtain the third amino acid pair encoding and the third MSA encoding. Then, the amino acid encoding is determined according to the third MSA encoding, and based on the amino acid encoding, the third amino acid pair encoding, and the initialized structure information, the three-dimensional atomic coordinates of the protein are predicted, and the complex conformation is constructed. The method of the embodiments of the present disclosure improves the accuracy of predicting the complex conformation of antigen and antibody and reduces the prediction cost of the complex conformation by obtaining the spatial structure information of amino acids and the interactions between amino acids. At the same time, it also provides a basis for drug design related to antigen and antibody, which helps to promote the development of antibody drug design.

[0082] Figure 4 It is a schematic flowchart of a method for obtaining the complex conformation of antigen and antibody provided by the embodiments of the present disclosure.

[0083] As Figure 4 shown, the method for obtaining the complex conformation of antigen and antibody may include:

[0084] S401, obtaining an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence.

[0085] S402, performing multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain.

[0086] S403, obtaining the complex conformation of antigen and antibody according to the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation.

[0087] The relevant content of steps S401 - S403 can be referred to the above embodiments and will not be elaborated here.

[0088] S404, obtaining the antibody properties of the reference antibody sequence according to the complex conformation, and performing expectation verification on the antibody properties of the reference antibody sequence.

[0089] In some implementations, the antibody properties of the reference antibody sequence are expected to be verified, which can verify whether the antibody properties meet the requirements. When the requirements are not met, the reference antibody sequence is optimized to improve the antibody properties of the reference antibody sequence.

[0090] In some implementations, antibody properties can be obtained based on the complex conformation. Optionally, the antibody properties include, but are not limited to: the affinity between the antibody and the antigen, the immunogenicity of the antibody, the solubility of the antibody, etc. Optionally, based on multiple pre-trained antibody property prediction models, the antibody properties of the complex conformation can be predicted respectively to obtain multiple antibody properties of the reference antibody sequence. Also, based on one antibody property prediction model, the antibody properties of the complex conformation can be predicted to jointly output multiple antibody properties of the reference antibody sequence.

[0091] Further, after obtaining the antibody properties of the reference antibody sequence, the property value of each type of antibody property can be determined. By comparing the property value with its corresponding judgment condition, the antibody properties of the reference antibody sequence can be verified to verify whether the antibody properties pass the expected verification.

[0092] In some implementations, the antibody properties of the reference antibody sequence at least include the affinity between the antibody and the antigen and the immunogenicity of the antibody. The judgment conditions for each type of antibody property of the reference antibody sequence can be determined based on the thresholds of the affinity between the antibody and the antigen and the immunogenicity of the antibody.

[0093] Optionally, the judgment conditions for each type of antibody property of the reference antibody sequence can be obtained based on the thresholds of each type of antibody property. Furthermore, the property value of each type of antibody property of the reference antibody sequence is compared with its respective judgment condition to verify the antibody properties of the reference antibody sequence.

[0094] Optionally, in response to the affinity property value being greater than the affinity threshold and the property value of the antibody immunogenicity being less than the antibody immunogenicity threshold, it is determined that the antibody properties pass the expected verification; otherwise, the antibody properties do not pass the expected verification.

[0095] Optionally, if the antibody properties of the reference antibody sequence pass the expected verification, the reference antibody sequence is determined to be the target antibody sequence of the antigen sequence. If the antibody properties of the reference antibody sequence do not pass the expected verification, the reference antibody sequence is mutated and the complex conformation is obtained again until the iterated antibody sequence is the target antibody sequence of the antigen sequence.

[0096] In some implementations, when the antibody properties of the reference antibody sequence fail to pass the desired verification, the reference antibody sequence can be mutated to improve its performance based on the reference antibody sequence, so as to optimize the antibody properties of the reference antibody sequence. Optionally, the mutable regions of the reference antibody sequence can be obtained, and the reference antibody sequence can be mutated based on the mutable regions to obtain at least one antibody sequence.

[0097] According to the method for obtaining the complex conformation of an antigen-antibody provided by an embodiment of the present disclosure, by obtaining an amino acid sequence and performing MSA on the amino acid sequence, the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain are obtained. Furthermore, by encoding and optimizing the amino acid sequence, the MSA sequence and the reference conformation, the residue coordinates of the protein can be predicted, and the complex conformation can be constructed, improving the accuracy of the antigen-antibody complex conformation prediction and reducing the prediction cost of the complex conformation. At the same time, it also provides a basis for drug design related to antigen-antibodies, contributing to the development of antibody drug design. Further, based on the complex conformation, antibody properties are predicted, and the optimization of the antibody sequence can be achieved, enabling the directional design and optimization of antibodies, thereby improving the design success rate.

[0098] As Figure 5 shown in the schematic flowchart of optimizing the antibody sequence. Figure 5 Taking the optimization of affinity as an example for illustration. By obtaining an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence, using the method of the embodiment of the present disclosure, the complex conformation of the antigen-antibody of the amino acid sequence is predicted to obtain the complex conformation of the antigen-antibody. Furthermore, the affinity of the complex conformation is predicted, and it is determined whether the affinity passes the desired verification. When it passes the desired verification, the reference antibody sequence is determined as the target antibody sequence; when it fails to pass the desired verification, the reference antibody sequence is mutated, and the above steps are re-executed.

[0099] Exemplarily, using the method provided by the embodiment of the present disclosure, the complex conformation of the antigen-antibody is obtained, and the affinity of the complex conformation is optimized.

[0100] 1. Input an amino acid sequence, including a reference antibody sequence and an antigen sequence, where the reference antibody sequence can be a wild-type antibody sequence.

[0101] 2. According to the input sequence, use the method provided by the embodiment of the present disclosure to predict the antigen-antibody complex conformation.

[0102] 3. Based on the conformational prediction results, use methods related to molecular dynamics to predict the affinity. For example, use Molecular Mechanics Generalized Born Surface Area (MMGBSA), or the Foldx model, etc. to predict the affinity.

[0103] 4. Determine whether the affinity meets the requirements according to the prediction results. For example, determine whether the affinity is greater than the set threshold. If the requirements are met, the optimized antibody sequence is obtained. If the requirements are not met, go to step 5.

[0104] 5. Mutate the antibody sequence and repeat the above steps. (For example, the hypervariable regions of the heavy chain or light chain of the antibody sequence can be mutated.)

[0105] Figure 6 It is a schematic flowchart of a method for obtaining the conformation of an antigen-antibody complex provided by an embodiment of the present disclosure.

[0106] As Figure 6 shown, the method for obtaining the conformation of the antigen-antibody complex may include:

[0107] S601. Obtain an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence.

[0108] S602. Perform MSA on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain.

[0109] S603. Obtain the first amino acid pair encoding between amino acids in the amino acid sequence.

[0110] S604. Obtain the first MSA encoding according to the characteristics of the MSA sequence and the characteristics of the amino acid sequence.

[0111] S605. Obtain the pairing characteristics of the reference conformation, encode the pairing characteristics to obtain the second encoding of the reference conformation, and splice the first amino acid pair encoding and the second encoding to obtain the second amino acid pair encoding.

[0112] S606. Obtain the angular characteristics of the reference conformation, encode the angular characteristics to obtain the first encoding of the reference conformation, and splice the first MSA encoding and the first encoding to obtain the second MSA encoding.

[0113] S607. Input the second amino acid pair encoding and the second MSA encoding into the Evoformer network, and perform information interaction on the second amino acid pair encoding and the second MSA encoding through the Evoformer network to obtain the optimized third amino acid pair encoding and third MSA encoding.

[0114] S608. Obtain the first row elements of the third MSA code and transform the first row elements to obtain amino acid codes.

[0115] S609. Perform coordinate prediction based on the amino acid codes, the third amino acid pair codes, and the initialized structural information to obtain the three-dimensional atomic coordinates of the protein.

[0116] S610. Determine the complex conformation of the antigen-antibody based on the three-dimensional atomic coordinates.

[0117] According to the method for obtaining the complex conformation of the antigen-antibody provided by the embodiments of the present disclosure, by obtaining the amino acid sequence and performing MSA on the amino acid sequence, the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain are obtained. Furthermore, by encoding and optimizing the amino acid sequence, the MSA sequence, and the reference conformation, the residue coordinates of the protein can be predicted, and the complex conformation can be constructed, improving the accuracy of the prediction of the antigen-antibody complex conformation and reducing the prediction cost of the complex conformation. At the same time, it also provides a basis for drug design related to antigen-antibodies, contributing to the development of antibody drug design.

[0118] Figure 7 The flowchart showing the process of obtaining the complex conformation. Obtain the amino acid sequence, perform MSA on the amino acid sequence to obtain the MSA sequence and the reference conformation, and input the amino acid sequence, the MSA sequence, and the reference conformation into the feature extraction module to respectively obtain the features of the amino acid sequence, the MSA sequence, and the reference conformation, and encode the features to obtain an encoded representation. Further, input the encoded representation into the geometric modeling module to optimize the encoded representation to obtain an optimized encoded representation. Then, the structural prediction module predicts the residue coordinates of the optimized encoded representation and constructs the complex conformation based on the residue coordinates, and finally outputs the complex conformation.

[0119] Figure 8To obtain a structural schematic diagram of the complex conformation. By obtaining the amino acid sequence and performing MSA on the amino acid sequence, the MSA sequence and the reference conformation are obtained. The first amino acid pair encoding is obtained by encoding the amino acid pair, and the first MSA encoding is obtained by encoding the MSA sequence. Based on the first MSA encoding and the first amino acid pair encoding, combined with the characteristics of the reference conformation, the first MSA encoding and the first amino acid pair encoding are spliced to obtain the second amino acid pair encoding and the second MSA encoding, and they are input into the Evoformer network together. The Evoformer network optimizes the second amino acid pair encoding and the second MSA encoding and outputs the third amino acid pair encoding and the third MSA encoding. The amino acid encoding can be obtained from the third MSA encoding, and the amino acid encoding and the third amino acid pair encoding are input into the IPA network for processing, and the amino acid encoding and the structural information are input into the CIN network for processing, and finally the target amino acid encoding and the target amino acid pair encoding are obtained. Combining the target amino acid encoding and the target amino acid pair encoding, the three-dimensional atomic coordinates of the protein are predicted and the complex conformation is constructed.

[0120] The embodiments of the present disclosure are applicable to scenarios where it is necessary to predict the complex conformation given the antigen-antibody sequences, including but not limited to the following scenarios:

[0121] Antigen-antibody docking: The embodiments of the present disclosure can be used to predict the antigen-antibody complex conformation, thereby assisting in antibody design.

[0122] Antigen-antibody conformation prediction: The embodiments of the present disclosure can be used for antigen-antibody complex conformation prediction, thereby assisting in docking surface analysis, affinity maturation, etc.

[0123] Antigen epitope discovery: The embodiments of the present disclosure can be used for antigen epitope analysis. By predicting the antigen-antibody complex conformation multiple times or once, and then analyzing the docking surface, antigen epitope information can be obtained.

[0124] Antibody optimization: The embodiments of the present disclosure can be used for antibody optimization. Predict the antigen-antibody complex conformation, and then based on the docking of the conformation and the analysis of residue characteristics, it can be used for optimizing the properties of the antibody, including but not limited to affinity, immunogenicity, activity, etc.

[0125] Disease mechanism research: The occurrence and development of many diseases are related to abnormal interactions between proteins. Protein-protein docking can help researchers understand the molecular mechanisms of these abnormal interactions, thereby providing new ideas for the diagnosis and treatment of diseases.

[0126] Corresponding to the method for obtaining the conformational antigen-antibody complex provided in the above several embodiments, an embodiment of the present disclosure further provides an apparatus for obtaining the conformational antigen-antibody complex. Since the apparatus for obtaining the conformational antigen-antibody complex provided in the embodiments of the present disclosure corresponds to the method for obtaining the conformational antigen-antibody complex provided in the above several embodiments, the implementation manners of the above method for obtaining the conformational antigen-antibody complex are also applicable to the apparatus for obtaining the conformational antigen-antibody complex provided in the embodiments of the present disclosure, and will not be described in detail in the following embodiments.

[0127] Figure 9 It is a schematic structural diagram of an apparatus for obtaining the conformational antigen-antibody complex provided in the embodiments of the present disclosure.

[0128] As Figure 9 shown, the apparatus 900 for obtaining the conformational antigen-antibody complex according to the embodiments of the present disclosure includes a first acquisition module 901, an MSA module 902, and a second acquisition module 903.

[0129] The first acquisition module 901 is used to acquire an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence.

[0130] The MSA module 902 is used to perform multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain.

[0131] The second acquisition module 903 is used to obtain the conformational antigen-antibody complex according to the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation.

[0132] In an embodiment of the present disclosure, the second acquisition module 903 is further used to: acquire the first amino acid pair encoding between amino acid pairs in the amino acid sequence; obtain the first MSA encoding according to the characteristics of the MSA sequence and the characteristics of the amino acid sequence; and obtain the conformational antigen-antibody complex according to the first amino acid pair encoding, the first MSA encoding, and the reference conformation.

[0133] In an embodiment of the present disclosure, the second acquisition module 903 is further used to: perform embedding layer encoding on the amino acid sequence to obtain the embedding encoding of the amino acid sequence; encode the index values of the amino acids in the amino acid sequence to obtain the position encoding; and fuse the embedding encoding and the position encoding to obtain the first amino acid pair encoding.

[0134] In one embodiment of the present disclosure, the second acquisition module 903 is further configured to: acquire the pairing features of the reference conformation, encode the pairing features to obtain the second encoding of the reference conformation, and splice the first amino acid pair encoding and the second encoding to obtain the second amino acid pair encoding; acquire the angular features of the reference conformation, encode the angular features to obtain the first encoding of the reference conformation, and splice the first MSA encoding and the first encoding to obtain the second MSA encoding; and acquire the complex conformation of the antigen-antibody according to the second amino acid pair encoding and the second MSA encoding.

[0135] In one embodiment of the present disclosure, the second acquisition module 903 is further configured to: input the second amino acid pair encoding and the second MSA encoding into the model backbone Evoformer network, perform information interaction on the second amino acid pair encoding and the second MSA encoding through the Evoformer network to obtain the optimized third amino acid pair encoding and third MSA encoding; and acquire the complex conformation of the antigen-antibody according to the third amino acid pair encoding and the third MSA encoding.

[0136] In one embodiment of the present disclosure, the second acquisition module 903 is further configured to: acquire the first row elements of the third MSA encoding, transform the first row elements to obtain the amino acid encoding; perform coordinate prediction according to the amino acid encoding, the third amino acid pair encoding, and the initialized structure information to obtain the three-dimensional atomic coordinates of the protein; and determine the complex conformation of the antigen-antibody according to the atomic coordinates.

[0137] In one embodiment of the present disclosure, the second acquisition module 903 is further configured to: input the amino acid encoding and the third amino acid pair encoding into the invariant attention network IPA network for processing to obtain the candidate amino acid encoding and the candidate amino acid pair encoding; input the amino acid encoding and the structure information into the chain-level information interaction network CIN network for processing, and update the candidate amino acid encoding and the candidate amino acid pair encoding based on the output of the CIN network to obtain the target amino acid encoding and the target amino acid pair encoding.

[0138] In one embodiment of the present disclosure, the device further includes: a verification module, configured to acquire the antibody attributes of the reference antibody sequence according to the complex conformation, and perform expectation verification on the antibody attributes of the reference antibody sequence.

[0139] In one embodiment of the present disclosure, the verification module is further configured to: if the antibody attributes of the reference antibody sequence pass the desired verification, determine that the reference antibody sequence is the target antibody sequence of the antigen sequence; if the antibody attributes of the reference antibody sequence do not pass the desired verification, mutate the reference antibody sequence and re-obtain the complex conformation until the iterated antibody sequence is the target antibody sequence of the antigen sequence.

[0140] In one embodiment of the present disclosure, the verification module is further configured to: obtain the determination conditions for each type of antibody attribute of the reference antibody sequence; compare the attribute values of each type of antibody attribute of the reference antibody sequence with their respective determination conditions to verify the antibody attributes of the reference antibody sequence.

[0141] In one embodiment of the present disclosure, the antibody attributes at least include the affinity between the antibody and the antigen and the immunogenicity of the antibody.

[0142] In one embodiment of the present disclosure, the verification module is further configured to: obtain the mutable regions of the reference antibody sequence and mutate the reference antibody sequence based on the mutable regions to obtain at least one antibody sequence.

[0143] According to the apparatus for obtaining the complex conformation of an antigen-antibody provided by the embodiments of the present disclosure, by obtaining an amino acid sequence and performing MSA on the amino acid sequence, the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain are obtained. Furthermore, by encoding and optimizing the amino acid sequence, the MSA sequence, and the reference conformation, the residue coordinates of the protein can be predicted, and the complex conformation can be constructed, improving the accuracy of predicting the complex conformation of the antigen-antibody and reducing the prediction cost of the complex conformation. At the same time, it also provides a basis for drug design related to antigen-antibodies, helping to promote the development of antibody drug design.

[0144] In the technical solution of the present disclosure, the acquisition, storage, and application of the user's personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0145] According to the embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0146] Figure 10FIG. shows a schematic block diagram of an exemplary electronic device 800 that can be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, for example, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementations of the present disclosure described and / or claimed herein.

[0147] As Figure 10 shown, the device 1000 includes a computing unit 1001 that can perform various appropriate actions and processes in accordance with computer programs / instructions stored in a read-only memory (ROM) 1002 or computer programs / instructions loaded from a storage unit 1006 into a random access memory (RAM) 1003. In the RAM 1003, various programs and data required for the operation of the device 1000 can also be stored. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0148] A plurality of components in the device 1000 are connected to the I / O interface 1005, including: an input unit 1006 such as a keyboard, a mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a magnetic disk, an optical disk, etc.; and a communication unit 1009, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1009 allows the device 1000 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0149] The computing unit 1001 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1001 executes the various methods and processes described above, such as the method for obtaining the complex conformation of antigen-antibody. For example, in some embodiments, the method for obtaining the complex conformation of antigen-antibody can be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as the storage unit 1006. In some embodiments, part or all of the computer program / instructions can be loaded and / or installed onto the device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program / instructions are loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the method for obtaining the complex conformation of antigen-antibody described above can be executed. Alternatively, in other embodiments, the computing unit 1001 can be configured to execute the method for obtaining the complex conformation of antigen-antibody by any other suitable means (e.g., by means of firmware).

[0150] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), systems-on-a-chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs / instructions, which can be executed and / or interpreted on a programmable system including at least one programmable processor, and the programmable processor can be a special or general programmable processor, receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0151] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program codes can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0152] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0153] To provide for interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0154] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of a communication network include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0155] A computer system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The client-server relationship is generated by computer programs / instructions running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

[0156] It should be understood that various forms of the processes shown above may be used, with steps reordered, added, or deleted. For example, the steps recited in the disclosure may be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and no limitations are imposed herein.

[0157] The above specific embodiments do not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

1. A method for obtaining the conformation of an antigen-antibody complex, wherein, The method includes: Obtaining an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence; Performing a multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain; Predicting the three-dimensional atomic coordinates of the protein based on the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation; Obtaining the complex conformation of the antigen-antibody based on the three-dimensional atomic coordinates of the protein.

2. The method according to claim 1, wherein The obtaining of the complex conformation of the antigen-antibody based on the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation includes: Obtaining the first amino acid pair encoding between amino acid pairs in the amino acid sequence; Obtaining the first MSA encoding based on the characteristics of the MSA sequence and the characteristics of the amino acid sequence; Predicting the three-dimensional atomic coordinates of the protein based on the first amino acid pair encoding, the first MSA encoding, and the reference conformation; Obtaining the complex conformation of the antigen-antibody based on the three-dimensional atomic coordinates of the protein.

3. The method according to claim 2, wherein, The obtaining of the first amino acid pair encoding between amino acid pairs in the amino acid sequence includes: Performing an embedding layer encoding on the amino acid sequence to obtain the embedding encoding of the amino acid sequence; Encoding the index values of the amino acids in the amino acid sequence to obtain the position encoding; Fusing the embedding encoding and the position encoding to obtain the first amino acid pair encoding.

4. The method according to claim 2, wherein The obtaining of the complex conformation of the antigen-antibody based on the first amino acid pair encoding, the first MSA encoding, and the reference conformation includes: Obtaining the pairing characteristics of the reference conformation, encoding the pairing characteristics to obtain the second encoding of the reference conformation, and splicing the first amino acid pair encoding and the second encoding to obtain the second amino acid pair encoding; Obtaining the angular characteristics of the reference conformation, encoding the angular characteristics to obtain the first encoding of the reference conformation, and splicing the first MSA encoding and the first encoding to obtain the second MSA encoding; Predicting the three-dimensional atomic coordinates of the protein based on the second amino acid pair encoding and the second MSA encoding, and obtaining the complex conformation of the antigen-antibody based on the three-dimensional atomic coordinates of the protein.

5. The method according to claim 4, wherein, The predicting of the three-dimensional atomic coordinates of the protein based on the second amino acid pair encoding and the second MSA encoding, and the obtaining of the complex conformation of the antigen-antibody based on the three-dimensional atomic coordinates of the protein includes: Inputting the second amino acid pair encoding and the second MSA encoding into the model backbone Evoformer network, and performing information interaction on the second amino acid pair encoding and the second MSA encoding through the Evoformer network to obtain the optimized third amino acid pair encoding and third MSA encoding; Obtaining the complex conformation of the antigen-antibody based on the third amino acid pair encoding and the third MSA encoding.

6. The method according to claim 5, wherein The obtaining of the complex conformation of the antigen-antibody based on the third amino acid pair encoding and the third MSA encoding includes: Obtain the first row elements encoded by the third MSA, and transform the first row elements to obtain amino acid codes; Perform coordinate prediction based on the amino acid codes, the third amino acid pair codes, and the initialized structural information to obtain the three-dimensional coordinates of the atoms; Determine the complex conformation of the antigen-antibody based on the three-dimensional coordinates of the atoms.

7. The method according to claim 6, wherein The performing coordinate prediction based on the amino acid codes, the third amino acid pair codes, and the initialized structural information to obtain the three-dimensional coordinates of the atoms includes: Input the amino acid codes and the third amino acid pair codes into the Invariant Point Attention (IPA) network for processing to obtain candidate amino acid codes and candidate amino acid pair codes; Input the amino acid codes and the structural information into the Chain-level Information Interaction Network (CIN) for processing, and update the candidate amino acid codes and the candidate amino acid pair codes based on the output of the CIN network to obtain target amino acid codes and target amino acid pair codes.

8. The method according to any one of claims 1-7, wherein, After obtaining the complex conformation of the antigen-antibody according to the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation, it further includes: Obtain the antibody properties of the reference antibody sequence according to the complex conformation, and perform expectation verification on the antibody properties of the reference antibody sequence.

9. The method according to claim 8, wherein The method further includes: If the antibody properties of the reference antibody sequence pass the expectation verification, determine that the reference antibody sequence is the target antibody sequence of the antigen sequence; If the antibody properties of the reference antibody sequence do not pass the expectation verification, mutate the reference antibody sequence and re-obtain the complex conformation until the iterated antibody sequence is the target antibody sequence of the antigen sequence.

10. The method according to claim 9, wherein, The performing expectation verification on the antibody properties of the reference antibody sequence includes: Obtain the determination conditions for each type of antibody property of the reference antibody sequence; Compare the property values of each type of antibody property of the reference antibody sequence with their respective judgment conditions to verify the antibody properties of the reference antibody sequence.

11. The method according to claim 10, wherein, The antibody properties at least include the affinity between the antibody and the antigen and the immunogenicity of the antibody.

12. The method according to claim 9, wherein, The mutating the reference antibody sequence includes: Obtain the mutable regions of the reference antibody sequence, and mutate the reference antibody sequence based on the mutable regions to obtain at least one antibody sequence.

13. An apparatus for obtaining the conformation of an antigen-antibody complex, wherein, The device includes: A first acquisition module, configured to acquire an amino acid sequence, where the amino acid sequence includes an antigen sequence and a reference antibody sequence; An MSA module, configured to perform multiple sequence alignment (MSA) on the amino acid sequence to obtain the MSA sequence of each protein chain and the reference conformation corresponding to the protein chain; A second acquisition module, configured to predict the three-dimensional coordinates of the atoms of the protein according to the amino acid sequence, the MSA sequence of the protein chain, and the reference conformation, and obtain the complex conformation of the antigen-antibody according to the three-dimensional coordinates of the atoms of the protein.

14. The apparatus according to claim 13, wherein, The second acquisition module is further configured to: Obtain the first amino acid pair codes between amino acid pairs in the amino acid sequence; Obtain a first MSA encoding based on the characteristics of the MSA sequence and the characteristics of the amino acid sequence. Predict the three-dimensional atomic coordinates of the protein according to the first amino acid pair encoding, the first MSA encoding, and the reference conformation, and obtain the complex conformation of the antigen-antibody according to the three-dimensional atomic coordinates of the protein.

15. The apparatus according to claim 14, wherein, The second acquisition module is further configured to: Perform an embedding layer encoding on the amino acid sequence to obtain an embedding encoding of the amino acid sequence; Encode the index values of the amino acids in the amino acid sequence to obtain a position encoding; Fuse the embedding encoding and the position encoding to obtain the first amino acid pair encoding.

16. The apparatus according to claim 14, wherein, The second acquisition module is further configured to: Obtain the pairing characteristics of the reference conformation, encode the pairing characteristics to obtain a second encoding of the reference conformation, and splice the first amino acid pair encoding and the second encoding to obtain a second amino acid pair encoding; Obtain the angular characteristics of the reference conformation, encode the angular characteristics to obtain a first encoding of the reference conformation, and splice the first MSA encoding and the first encoding to obtain a second MSA encoding; Predict the three-dimensional atomic coordinates of the protein according to the second amino acid pair encoding and the second MSA encoding, and obtain the complex conformation of the antigen-antibody according to the three-dimensional atomic coordinates of the protein.

17. The apparatus according to claim 16, wherein, The second acquisition module is further configured to: Input the second amino acid pair encoding and the second MSA encoding into the model backbone Evoformer network, and perform information interaction on the second amino acid pair encoding and the second MSA encoding through the Evoformer network to obtain an optimized third amino acid pair encoding and a third MSA encoding; Predict the three-dimensional atomic coordinates of the protein according to the third amino acid pair encoding and the third MSA encoding, and obtain the complex conformation of the antigen-antibody according to the three-dimensional atomic coordinates of the protein.

18. The apparatus according to claim 17, wherein, The second acquisition module is further configured to: Obtain the first row elements of the third MSA encoding, and transform the first row elements to obtain an amino acid encoding; Perform coordinate prediction according to the amino acid encoding, the third amino acid pair encoding, and the initialized structural information to obtain the three-dimensional atomic coordinates of the protein; Determine the complex conformation of the antigen-antibody according to the three-dimensional atomic coordinates.

19. The apparatus according to claim 18, wherein, The second acquisition module is further configured to: Input the amino acid encoding and the third amino acid pair encoding into the Invariant Point Attention (IPA) network for processing to obtain a candidate amino acid encoding and a candidate amino acid pair encoding; Input the amino acid encoding and the structural information into the Chain-Level Information Network (CIN) for processing, and update the candidate amino acid encoding and the candidate amino acid pair encoding based on the output of the CIN network to obtain a target amino acid encoding and a target amino acid pair encoding.

20. The apparatus according to any one of claims 13-19, wherein, The apparatus further includes: A verification module, configured to obtain the antibody attributes of the reference antibody sequence according to the complex conformation, and perform an expectation verification on the antibody attributes of the reference antibody sequence.

21. The device according to claim 20, wherein, The verification module is further configured to: If the antibody properties of the reference antibody sequence pass the desired verification, determine that the reference antibody sequence is the target antibody sequence of the antigen sequence; If the antibody properties of the reference antibody sequence do not pass the desired verification, mutate the reference antibody sequence and re-obtain the complex conformation until the iterated antibody sequence is the target antibody sequence of the antigen sequence.

22. The apparatus according to claim 21, wherein, The verification module is further configured to: Obtain the determination conditions for each type of antibody property of the reference antibody sequence; Compare the property values of each type of antibody property of the reference antibody sequence with their respective judgment conditions to verify the antibody properties of the reference antibody sequence.

23. The apparatus according to claim 22, wherein, The antibody properties at least include the affinity between the antibody and the antigen and the immunogenicity of the antibody.

24. The apparatus according to claim 21, wherein, The verification module is further configured to: Obtain the mutable region of the reference antibody sequence and mutate the reference antibody sequence based on the mutable region to obtain at least one antibody sequence.

25. 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 to enable the at least one processor to execute the method according to any one of claims 1-12.

26. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to execute the method according to any one of claims 1-12.

27. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, the method according to any one of claims 1-12 is implemented.

Images