Methods, apparatuses, and devices for antibody design and storage media

Abstract

The application relates to a method, device and equipment for designing an antibody and a storage medium. The method comprises the following steps: screening a template sequence from an antibody sequence library according to a sequence of a CDRH3 region of an antibody to be designed; replacing the sequence of the CDRH3 region of the template sequence with the sequence of the CDRH3 region of the antibody to be designed as a candidate antibody sequence; predicting a conformation of the candidate antibody according to the candidate antibody sequence; optimizing the conformation of the CDRH3 region of the candidate antibody by adding a constraint condition based on a preset algorithm; docking the CDRH3 regions of a plurality of optimized antibodies with an antigen target epitope to obtain a plurality of complex conformations and sequences of the optimized antibody-antigen; and optimizing a binding interface of the complex conformation of the optimized antibody-antigen to obtain a plurality of target antibody conformations and sequences capable of binding the antigen target epitope. The scheme provided in the application can obtain a complex conformation of the antibody binding the antigen target epitope as a whole, and can improve the success rate of epitope antibody design.

Description

Technical Field

[0001] This application relates to the field of bioinformatics, and in particular to antibody design methods, devices and equipment, and storage media. Background Technology

[0002] In the field of antibody applications, antibodies targeting specific epitopes (epitope-specific antibodies) have significant advantages in research, diagnosis, and treatment. Methods for designing epitope-specific antibodies through computational simulation include sequence-based antibody design and structure-based antibody design. The main difference between these two methods is that structure-based antibody design can simulate the structure of the candidate antibody molecule binding to the epitope, generating an antigen-antibody complex structure, while sequence-based antibody design does not simulate the antigen-antibody complex structure of the candidate antibody molecule. Therefore, structure-based antibody design theoretically has a higher success rate.

[0003] In related technologies, some AI models can design conformations of antibody CDRH3 binding to antigen target epitopes, referred to as CDRH3-epitope conformations.

[0004] However, the above methods only reach the level of generating CDRH3-epitope conformations, and can only produce the sequence of antibody CDRH3, but cannot obtain the conformation of the antibody binding the antigen target epitope as a whole. Summary of the Invention

[0005] To address or partially address the problems existing in related technologies, this application provides an antibody design method, apparatus, device, and storage medium, which can obtain a complex conformation of the antibody binding to the target antigen epitope based on the binding conformation of the antibody CDRH3-antigen epitope, and obtain multiple antibody conformations and sequences that can bind antigens by optimizing the complex conformation, thereby improving the success rate of antibody design.

[0006] The first aspect of this application provides a method for designing antibodies, including:

[0007] Obtain the sequence of the CDRH3 region of the antibody to be designed and the binding conformation of the antibody to the target epitope CDRH3-antigen;

[0008] Based on the sequence of the CDRH3 region of the antibody to be designed, an antibody sequence that meets the first preset screening condition is selected from the antibody sequence library as a template sequence;

[0009] The CDRH3 region of the template sequence is replaced with the CDRH3 region of the antibody to be designed, and this is used as a candidate antibody sequence.

[0010] Based on the candidate antibody sequence, the conformation of the candidate antibody is predicted;

[0011] Based on a preset algorithm, the conformation of the CDRH3 region of the candidate antibody is optimized by adding constraints to obtain multiple optimized antibody conformations. The constraints can be set so that the conformations of the CDRH3 region of the multiple optimized antibodies satisfy the binding characteristics of the CDRH3-antigen target epitope binding conformation of the antibody to be designed and the conformational diversity characteristics of the CDRH3 region of the candidate antibody.

[0012] Based on the binding conformation of the antibody CDRH3-antigen target epitope to be designed, the CDRH3 regions of multiple optimized antibodies are docked with the antigen target epitope to obtain multiple optimized antibody-antigen complex conformations and sequences.

[0013] Based on the multiple optimized antibody-antigen complex conformations and the interactions between the optimized antibody and the antigen, optimized antibody-antigen complex conformations and sequences that meet the second preset screening conditions are screened.

[0014] The binding interface of the screened optimized antibody-antigen complex was structurally and sequentially optimized to obtain the structures and sequences of multiple target antibodies that can bind to antigen target epitopes.

[0015] As an optional embodiment, obtaining the sequence of the CDRH3 region of the antibody to be designed and the binding conformation of the antibody to be designed CDRH3-antigen target epitope includes:

[0016] Based on the conformation of the antigen and the target epitope, and the preset sequence length of the CDRH3 region of the antibody to be designed, generate a number of sequences of the CDRH3 region of the antibody to be designed corresponding to the sequence length;

[0017] Select a predetermined number of sequences from all the CDRH3 regions of the antibodies to be designed;

[0018] Based on the selected sequence of each CDRH3 region of the antibody to be designed, multiple binding conformations of the antibody to be designed CDRH3-antigen target epitopes are generated.

[0019] As an optional embodiment, the step of selecting antibody sequences that meet the first preset screening conditions from the antibody sequence library as template sequences based on the sequence of the CDRH3 region of the antibody to be designed includes:

[0020] Based on the sequence length of the CDRH3 region of the antibody to be designed, multiple antibody sequences with the same CDRH3 region sequence length as the antibody to be designed are selected from the antibody sequence library and used as candidate antibody sequences.

[0021] By comparing the number of different amino acids contained in the CDRH3 region of the candidate antibody sequence with the number of different amino acids contained in the CDRH3 region of the antibody to be designed, the candidate antibody sequence with the fewest different amino acids is selected as the template sequence.

[0022] As an optional embodiment, the second preset filtering conditions include:

[0023] In the optimized antibody-antigen complex conformation, the conformations of the optimized antibody and the antigen do not conflict, and in the optimized antibody-antigen complex conformation, the interaction between the optimized antibody and the antigen in regions other than CDRH3 is not lower than a threshold.

[0024] As an optional embodiment, the method further includes:

[0025] Based on the binding conformation of the antibody CDRH3-antigen target epitope to be designed, the binding characteristics of the binding conformation of the antibody CDRH3-antigen target epitope to be designed are extracted. The binding characteristics include the interaction residue pairs between the antibody CDRH3 region to be designed and the antigen target epitope, as well as the dihedral angles of the corresponding residue sites on the antibody CDRH3 region to be designed.

[0026] As an optional embodiment, the step of extracting the binding characteristics of the binding conformation of the target epitope of the antibody CDRH3-antigen based on the binding conformation of the target epitope of the antibody to be designed includes:

[0027] Based on the binding conformation of the antibody CDRH3 to the target epitope of the antigen, calculate the distance between the CDRH3 region of the antibody to be designed and the residue pair on the target epitope of the antigen.

[0028] The sites where residue pairs with a distance less than a preset value are located are taken as interacting residue pairs, and the dihedral angles of the interacting residue pairs and the residue sites located on the CDRH3 region of the antibody to be designed are taken as the binding characteristics of the binding conformation of the antibody to be designed CDRH3-antigen target epitope.

[0029] As an optional embodiment, the constraint conditions include at least one of condition one, condition two, condition three, and condition four, wherein,

[0030] Condition one is: based on the residue sites on the CDRH3 region of the antibody to be designed that interact with the target epitope of the antigen, constrain the dihedral angle of the corresponding residue sites on the CDRH3 region of the candidate antibody;

[0031] The second condition is: constraining the distance of hydrogen bonds in conserved regions within the CDRH3 region of the candidate antibody;

[0032] The third condition is: based on the conformational clustering results of the antibody CDRH3 region in the existing database, according to the sequence length of the CDRH3 region of the antibody to be designed, the corresponding conformationally conserved residue sites in each clustering result are obtained, and the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody are constrained after excluding the constrained residue sites in the first condition.

[0033] Condition four is: after excluding the constrained residue sites in conditions one and three, the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody are constrained according to a set method.

[0034] As an optional embodiment, the step of docking the CDRH3 regions of multiple optimized antibodies with the target epitope of the antigen according to the binding conformation of the antibody CDRH3-antigen, to obtain multiple optimized antibody-antigen complex conformations and sequences, includes:

[0035] Based on the binding conformation of the target epitope of the antibody to be designed CDRH3-antigen, the positions of multiple optimized antibodies are adjusted until the positions of the CDRH3 regions of the optimized antibodies and the CDRH3 regions of the antibody to be designed are basically superimposed, thus obtaining multiple optimized antibody-antigen complex conformations and sequences.

[0036] A second aspect of this application provides an antibody design device, comprising:

[0037] The acquisition module is used to acquire the sequence of the CDRH3 region of the antibody to be designed and the binding conformation of the CDRH3-antigen target epitope of the antibody to be designed;

[0038] The first screening module is used to select antibody sequences that meet the first preset screening conditions from the antibody sequence library as template sequences based on the sequence of the CDRH3 region of the antibody to be designed.

[0039] The replacement module is used to replace the CDRH3 region of the template sequence with the CDRH3 region of the antibody to be designed, and use it as a candidate antibody sequence.

[0040] A prediction module is used to predict the conformation of the candidate antibody based on the candidate antibody sequence;

[0041] The first optimization module is used to optimize the conformation of the CDRH3 region of the candidate antibody by adding constraints based on a preset algorithm, thereby obtaining the conformations of multiple optimized antibodies. The constraints are set so that the conformations of the CDRH3 region of the multiple optimized antibodies satisfy the binding characteristics of the binding conformation of the CDRH3-antigen target epitope of the antibody to be designed and the conformational diversity characteristics of the CDRH3 region of the candidate antibody.

[0042] The docking module is used to dock the CDRH3 region of multiple optimized antibodies with the target epitope of the antigen according to the binding conformation of the antibody CDRH3-antigen target epitope to be designed, so as to obtain multiple optimized antibody-antigen complex conformations and sequences.

[0043] The second screening module is used to screen optimized antibody-antigen complex conformations and sequences that meet the second preset screening conditions based on the multiple optimized antibody-antigen complex conformations and the interaction between the optimized antibody and the antigen.

[0044] The second optimization module is used to optimize the structure and sequence of the binding interface of the screened optimized antibody-antigen complex to obtain the structure and sequence of multiple target antibodies that can bind to antigen target epitopes.

[0045] A third aspect of this application provides an electronic device, comprising:

[0046] Processor; and

[0047] A memory that stores executable code, which, when executed by the processor, causes the processor to perform the method described above.

[0048] A fourth aspect of this application provides a computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the method described above.

[0049] The technical solution provided in this application may include the following beneficial effects:

[0050] This application selects a template sequence from an antibody sequence library based on the CDRH3 region sequence of the antibody to be designed. The CDRH3 region of the template sequence is substantially the same as that of the antibody to be designed, which increases the probability of correct antibody expression. Then, the CDRH3 region sequence of the template sequence is replaced with the CDRH3 region sequence of the antibody to be designed, and this is used as a candidate antibody sequence. Based on the candidate antibody sequence, the conformation of the candidate antibody is predicted. Then, based on a preset algorithm, the conformation of the CDRH3 region of the candidate antibody is optimized by adding constraints, so that the conformations of the CDRH3 regions of the multiple optimized antibodies satisfy the binding characteristics of the CDRH3-antigen target epitope binding conformation of the antibody to be designed and the diversity characteristics of the CDRH3 region of the antibody to be designed. This results in optimized antibody conformations that possess both the binding characteristics of the antibody to be designed CDRH3 and the target epitope of the antigen, and also satisfy the conformational diversity characteristics of the CDRH3 region of the candidate antibody. Finally, the CDRH3 regions of the multiple optimized antibodies are docked with the target epitope of the antigen to obtain multiple optimized antibody-antigen complex conformations and sequences. In the above steps, the desired antibody conformation is obtained by combining the binding characteristics of the antibody to be designed CDRH3 with the target epitope of the antigen. Based on the binding characteristics of the antigen target epitope and the optimized antibody conformation that satisfies the conformational diversity characteristics of the CDRH3 region of the candidate antibody, the interaction between the optimized antibody and the antigen is further determined by optimizing the conformation of the complex formed by the antibody-antigen binding. In addition to determining the interaction between the CDRH3 region of the optimized antibody and the antigen target epitope, the interaction between other regions of the optimized antibody (e.g., CDRL1, CDRL2, CDRL3, CDRH1, CDRH2, FWRH, FWRL) and the antigen containing the target epitope and other epitopes can also be determined. Next, based on the multiple optimized antibody-antigen complex conformations and the interactions between the optimized antibody and the antigen, optimized antibody-antigen complex conformations and sequences that meet the second preset screening conditions are screened. This removes complexes with conflicting binding conformations that prevent binding and / or weak interactions. Among the remaining complexes, the optimized antibody and antigen conformations do not conflict and can bind, with strong interactions and a high probability and degree of binding. Finally, the binding interface of the remaining optimized antibody-antigen complex conformations is structurally and sequentially optimized to obtain multiple target antibody conformations and sequences that can bind the antigen target epitope. Therefore, this application obtains the complete conformation of antibody-antigen binding based on the local conformation of antibody CDRH3 binding to the target epitope, thereby obtaining the structure and sequence of a complete antibody that can bind to the target epitope, which greatly increases the success rate of antibody design.

[0051] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0052] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.

[0053] Figure 1 This is a flowchart illustrating the antibody design method shown in the embodiments of this application;

[0054] Figure 2 This is another schematic flowchart illustrating the antibody design method shown in the embodiments of this application;

[0055] Figure 3 This is a schematic flowchart illustrating the antibody design method in the embodiments of this application;

[0056] Figure 4 This is a schematic diagram of the antibody design device shown in the embodiments of this application;

[0057] Figure 5 This is a schematic diagram of the structure of an electronic device shown in an embodiment of this application. Detailed Implementation

[0058] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0059] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0060] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0061] In related technologies, some AI models can design conformations of antibody CDRH3 binding to antigen target epitopes, referred to as CDRH3-epitope conformations.

[0062] However, the above methods only reach the level of generating CDRH3-epitope conformations, and can only produce the sequence of antibody CDRH3, but cannot obtain the conformation of the antibody binding the antigen target epitope as a whole.

[0063] To address the aforementioned issues, this application provides an antibody design method that can obtain a complex conformation of the antibody binding to the target antigen epitope based on the binding conformation of the antibody CDRH3-antigen epitope. Furthermore, by optimizing the complex conformation, multiple antibody conformations and sequences capable of binding antigens can be obtained, thereby improving the success rate of antibody design.

[0064] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0065] Figure 1 This is a flowchart illustrating the antibody design method shown in the embodiments of this application.

[0066] See Figure 1 This application provides an antibody design method, including steps S1 to S8:

[0067] S1: Obtain the sequence of the CDRH3 region of the antibody to be designed and the binding conformation of the CDRH3-antigen target epitope of the antibody to be designed.

[0068] In some embodiments, the present application embodiments may obtain the sequence of the CDRH3 region of the antibody to be designed and the binding conformation of the CDRH3 region of the antibody to be designed with the antigen target epitope in the following manner: generating a number of CDRH3 regions of the antibody to be designed corresponding to the sequence length based on the conformation of the antigen and the target epitope, and a preset sequence length of the CDRH3 region of the antibody to be designed; selecting a preset number of sequences from all the CDRH3 regions of the antibody to be designed; and generating multiple binding conformations of the CDRH3 region of the antibody to be designed with the antigen target epitope based on each selected CDRH3 region sequence of the antibody to be designed.

[0069] The antibody to be designed can refer to an antibody whose CDRH3 region can bind to the target epitope of the antigen; the target epitope is the site on the antigen molecule that specifically binds to the antibody and determines the antigen's specificity. The target epitope can be an epitope region at a specified location on the antigen; the antibody CDRH3 region plays a major role in determining the antibody's specificity and participates in recognizing and binding to antigens related to its specificity; the binding conformation of the antibody CDRH3 to the target epitope represents the conformation of the complex formed by the binding of the antibody CDRH3 region to the target epitope.

[0070] Additionally, it should be noted that the conformation in the embodiments of this application can refer to a 3D coordinate of the antigen / antibody sequence. One sequence can correspond to multiple conformations, and the conformation can be used as a basis for judging whether the antigen and antibody bind or whether the antigen-antibody complex is stable.

[0071] S2: Based on the sequence of the CDRH3 region of the antibody to be designed, select antibody sequences that meet the first preset screening conditions from the antibody sequence library as template sequences.

[0072] In some embodiments, the template sequence can be selected as follows: based on the sequence length of the CDRH3 region of the antibody to be designed, multiple antibody sequences with the same CDRH3 region length as the CDRH3 region of the antibody to be designed are selected from the antibody sequence library and used as candidate antibody sequences; the number of different amino acids contained in the CDRH3 region of the candidate antibody sequences is compared with the number of different amino acids contained in the CDRH3 region of the antibody to be designed, and the candidate antibody sequence with the fewest different amino acids is selected as the template sequence. It can be understood that the first preset screening condition can be: selecting the antibody sequence with the same CDRH3 region length as the CDRH3 region of the antibody to be designed and the fewest different amino acids contained in the CDRH3 region from the antibody sequence library as the template sequence.

[0073] The CDRH3 region of the template sequence selected in step S2 has the same sequence length as the CDRH3 region of the antibody to be designed, and contains basically the same amino acids. Therefore, it can be considered that the CDRH3 region of the template sequence is basically the same as the CDRH3 region of the antibody to be designed. Thus, the non-CDRH3 region sequence of the template sequence can be used as the non-CDRH3 region sequence of the antibody to be designed.

[0074] It should also be noted that the antibody sequence libraries in the embodiments of this application include, but are not limited to, the scFv antibody sequence library and the Fab antibody sequence library. The sources of the antibody sequence libraries are not limited to these, and this application does not limit them.

[0075] S3: Replace the CDRH3 region of the template sequence with the CDRH3 region of the antibody to be designed, and use it as a candidate antibody sequence.

[0076] In this embodiment, among the six CDR regions of the antibody to be designed, the CDRH3 region exhibits greater specificity in binding to the antigen than the other CDR regions. Therefore, this embodiment focuses on designing this most critical region.

[0077] In this embodiment, the CDRH3 region of the antibody to be designed is the CDR3 region of the heavy chain sequence. The antibody sequence library contains both heavy chain (IGHV) and light chain (IGKV / IGLV) sequences. Therefore, during replacement, the CDR3 region of the heavy chain sequence in the template sequence is replaced with the CDRH3 region of the antibody to be designed. It should be noted that the light chain sequence of the antibody to be designed can be set according to business needs or experience, and this application does not limit this. The candidate antibody sequence obtained through step S3 has both a complete sequence and retains the CDRH3 region sequence of the antibody to be designed.

[0078] S4: Based on the candidate antibody sequence, predict the conformation of the candidate antibody.

[0079] The embodiments of this application may use algorithms such as AlphaFold and Immubuilder to predict the conformation of candidate antibodies based on the candidate antibody sequence, and this application does not limit this.

[0080] S5: Based on the preset algorithm, the conformation of the CDRH3 region of the candidate antibody is optimized by adding constraints to obtain the conformation of multiple optimized antibodies. The constraint settings can make the conformation of the CDRH3 region of multiple optimized antibodies meet the binding characteristics of the CDRH3-antigen target epitope binding conformation of the antibody to be designed and the conformational diversity characteristics of the CDRH3 region of the candidate antibody.

[0081] In some embodiments, the constraints of this application may include at least one of condition one, condition two, condition three, and condition four, wherein condition one is: constraining the dihedral angles of corresponding residue sites on the CDRH3 region of the candidate antibody based on the residue sites on the CDRH3 region of the antibody to be designed that interact with the target epitope of the antigen; condition two is: constraining the distance of hydrogen bonds in conserved regions within the CDRH3 region of the candidate antibody; condition three is: based on the conformational clustering results of antibody CDRH3 regions in existing databases, obtaining the corresponding conformationally conserved residue sites in each clustering result according to the sequence length of the CDRH3 region of the antibody to be designed, and constraining the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody after excluding the residue sites already constrained in condition one; condition four is: constraining the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody according to a set method after excluding the residue sites already constrained in conditions one and three.

[0082] In this embodiment, step S5 optimizes the conformation of the CDRH3 region of the candidate antibody by adding constraints based on a preset algorithm. This results in multiple optimized antibody conformations that satisfy the binding characteristics of the CDRH3-antigen target epitope binding conformation of the antibody to be designed and the conformational diversity characteristics of the CDRH3 region of the candidate antibody. As a result, an optimized antibody conformation is obtained that has both the binding characteristics of the CDRH3 of the antibody to be designed and the antigen target epitope binding conformation and the conformational diversity characteristics of the CDRH3 region of the candidate antibody.

[0083] The preset algorithm in the embodiments of this application can be a protein molecule simulation algorithm, and this application does not limit it.

[0084] S6: Based on the binding conformation of the antibody CDRH3-antigen target epitope to be designed, the CDRH3 region of multiple optimized antibodies is docked with the antigen target epitope to obtain multiple optimized antibody-antigen complex conformations and sequences.

[0085] In some embodiments, the present application embodiments can obtain multiple optimized antibody-antigen complex conformations and sequences by: adjusting the positions of multiple optimized antibodies according to the binding conformation of the target epitope of the antibody to be designed CDRH3-antigen, until the position of the CDRH3 region of the optimized antibody substantially overlaps with the position of the CDRH3 region of the antibody to be designed, thereby obtaining multiple optimized antibody-antigen complex conformations and sequences.

[0086] In this embodiment, step S6 involves docking the CDRH3 regions of multiple optimized antibodies with the target epitopes of the antigen to obtain multiple optimized antibody-antigen complex conformations and sequences. Based on the optimized antibody conformations obtained in step S5, which possess both the binding characteristics of the CDRH3 region of the antibody to be designed and the target epitopes of the antigen, and satisfy the conformational diversity characteristics of the CDRH3 region of the candidate antibody, the interaction between the optimized antibody and the antigen is further determined by optimizing the conformation of the complex formed by the antibody-antigen binding. Moreover, in addition to determining the interaction between the CDRH3 region of the optimized antibody and the target epitopes of the antigen, the interaction between other regions of the optimized antibody (e.g., CDRL1, CDRL2, CDRL3, CDRH1, CDRH2, FWRH, FWRL) and the antigen containing the target epitopes and other epitopes can also be determined.

[0087] S7: Based on multiple optimized antibody-antigen complex conformations and optimized antibody-antigen interactions, screen for optimized antibody-antigen complex conformations and sequences that meet the second preset screening criteria.

[0088] In some embodiments, the second preset screening condition in the embodiments of this application may be: the conformations of the optimized antibody and antigen in the optimized antibody-antigen complex conformation do not conflict, and the interaction between the optimized antibody and antigen in the region outside CDRH3 in the optimized antibody-antigen complex conformation is not lower than a threshold.

[0089] In this embodiment of the application, the binding conformation of the antibody and antigen in the antibody-antigen complex is optimized, as well as the interaction between the antibody and antigen is optimized. Complexes with conflicting binding conformations that prevent binding and have weak interactions are removed. The remaining complexes have optimized antibody and antigen conformations that do not conflict and can bind, and have strong interactions with a high probability and degree of binding.

[0090] S8: The binding interface of the optimized antibody-antigen complex conformation that has passed screening is structurally and sequentially optimized to obtain the structure and sequence of multiple target antibodies that can bind to the target epitopes of the antigen.

[0091] This application embodiment optimizes the structure and sequence of the binding interface of the screened optimized antibody-antigen complex conformation, thereby structurally enhancing the interaction of the antibody-antigen binding interface and increasing the success rate of antibody design.

[0092] The embodiments of this application can optimize the binding interface of the remaining antibody-antigen complex conformation using interface optimization algorithms such as protein molecule simulation algorithms and ProteinMPNN, and this application does not limit this.

[0093] As can be seen from this example, in this embodiment, based on the sequence of the CDRH3 region of the antibody to be designed, a template sequence is selected from the antibody sequence library. The CDRH3 region of the template sequence is basically the same as that of the CDRH3 region of the antibody to be designed. This step can improve the probability of correct expression of the designed antibody. Then, the sequence of the CDRH3 region of the template sequence is replaced with the sequence of the CDRH3 region of the antibody to be designed, and used as a candidate antibody sequence. Then, based on the candidate antibody sequence, the conformation of the candidate antibody is predicted. Then, based on a preset algorithm, the conformation of the CDRH3 region of the candidate antibody is optimized by adding constraints, so that the conformations of the CDRH3 regions of the multiple optimized antibodies satisfy the binding characteristics of the CDRH3-antigen target epitope binding conformation of the antibody to be designed and the conformational diversity characteristics of the CDRH3 region of the candidate antibody. Thus, an optimized antibody conformation that has both the binding characteristics of the CDRH3 of the antibody to be designed and the antigen target epitope binding conformation and the conformational diversity characteristics of the CDRH3 region of the candidate antibody is obtained. Then, the CDRH3 regions of the multiple optimized antibodies are docked with the antigen target epitope to obtain multiple optimized antibody-antigen complex conformations and sequences. In the above steps, an optimized antibody conformation that has both the binding characteristics of the antibody to be designed and the antigen target epitope binding conformation and the conformational diversity characteristics of the CDRH3 region of the candidate antibody is obtained. Based on the binding characteristics of CDRH3 to the target epitope of the antigen and the optimized antibody conformation that satisfies the diversity characteristics of the CDRH3 region of the candidate antibody, the interaction between the optimized antibody and the antigen is further determined by optimizing the conformation of the complex formed by the antibody-antigen binding. In addition to determining the interaction between the CDRH3 region of the optimized antibody and the target epitope of the antigen, the interaction between other regions of the optimized antibody (such as CDRL1, CDRL2, CDRL3, CDRH1, CDRH2, FWRH, FWRL) and the antigen containing the target epitope and other epitopes can also be determined. Next, based on the multiple optimized antibody-antigen complex conformations and the interaction between the optimized antibody and the antigen, optimized antibody-antigen complex conformations and sequences that meet the second preset screening conditions are screened. This removes complexes with conflicting binding conformations that prevent binding and / or weak interactions. Among the remaining complexes, the optimized antibody and antigen conformations do not conflict and can bind, and the interaction is strong with a high probability and degree of binding. Finally, the binding interface of the remaining optimized antibody-antigen complex conformations is structurally and sequentially optimized to obtain multiple target antibody conformations and sequences that can bind to the target epitope of the antigen. Therefore, the embodiments of this application obtain the complete conformation of antibody-antigen binding based on the local conformation of antibody CDRH3 binding to the target epitope, thereby obtaining the structure and sequence of a complete antibody that can bind to the target epitope, which greatly increases the success rate of antibody design.

[0094] Figure 2 This is another schematic flowchart illustrating the antibody generation method shown in the embodiments of this application; Figure 3 This is a flowchart illustrating the antibody generation method in the embodiments of this application.

[0095] See Figure 2 and Figure 3 This application provides a method for generating antibodies, comprising the following steps:

[0096] S10: Based on the conformation of the antigen and the target epitope, and the preset sequence length of the CDRH3 region of the antibody to be designed, generate a number of CDRH3 regions of the antibody to be designed corresponding to the sequence length of the CDRH3 region of the antibody to be designed.

[0097] Steps S10 to S13 of this embodiment can be implemented using, for example, the HERN model. In S10, the conformation of the antigen and the target epitope (which may be an epitope at a specified position on the input antigen), as well as the preset sequence length of the CDRH3 region of the antibody to be designed, are used as input information for the model. The model generates a number of CDRH3 region sequences of the antibody to be designed corresponding to the sequence length based on the input information. If the conformation of the antigen and the target epitope are the same, but the preset sequence length of the CDRH3 region of the antibody to be designed is different, the number of CDRH3 region sequences of the antibody to be designed generated will be different. For example, if the preset sequence length of the CDRH3 region of the antibody to be designed is 10, hundreds of different CDRH3 region sequences of the antibody to be designed can be generated; if the preset sequence length of the CDRH3 region of the antibody to be designed is 20, thousands of different CDRH3 region sequences of the antibody to be designed can be generated.

[0098] In this embodiment of the application, the sequence can be numbered using the IMGT numbering method. The sequence length can refer to the number of amino acid numbers. For example, if the sequence length of the CDRH3 region of the antibody to be designed is 10, it means that the number of amino acid numbers in the CDRH3 region of the antibody to be designed is 10.

[0099] S11: Select a predetermined number of sequences from all CDRH3 regions of the antibody to be designed.

[0100] Since there are a large number of sequences in the CDRH3 region of the antibody to be designed generated according to the preset sequence length, the embodiments of this application can sort the generated sequences by, for example, the perplexity index and select the top k sequences.

[0101] Furthermore, the sorting method and selection quantity in the embodiments of this application can be set according to business needs or experience, and this application does not limit them.

[0102] S12: Based on the selected CDRH3 region sequence of each antibody to be designed, generate multiple binding conformations of CDRH3-antigen epitopes for the antibody to be designed.

[0103] The HERN model can generate multiple corresponding antibody CDRH3-antigen epitope binding conformations based on the sequence of each CDRH3 region of the antibody to be designed. Figure 3 (Represented by the CDRH3-epitope conformation in Chinese).

[0104] S13: Based on the binding conformation of the antibody CDRH3 to the target epitope of the antigen, extract the binding characteristics of the binding conformation of the antibody CDRH3 to the target epitope of the antigen. The binding characteristics include the interaction residue pairs between the CDRH3 region of the antibody and the target epitope of the antigen, as well as the dihedral angles of the corresponding residue sites on the CDRH3 region of the antibody.

[0105] In some embodiments, the binding features of this application can be extracted in the following manner: based on the binding conformation of the antibody CDRH3 to be designed and the target epitope of the antigen, the distance between the CDRH3 region of the antibody to be designed and the residue pair on the target epitope of the antigen is calculated; the sites where the residue pair with a distance less than a preset distance is located are taken as the interacting residue site pair, and the dihedral angle of the interacting residue site pair and the residue site located on the CDRH3 region of the antibody to be designed in the interacting residue site pair is taken as the binding feature of the binding conformation of the antibody CDRH3 to the target epitope of the antigen.

[0106] For example, each residue on the CDRH3 region of the antibody to be designed could be paired with each residue on the target epitope of the antigen to form a residue pair. The distance between the atoms contained in each residue pair could be calculated, and the minimum distance between atoms could be less than [the specified value]. The residue pairs are used as interacting residue pairs. The phi angle and psi angle of the residues located on the CDRH3 region of the antibody to be designed are used as the binding characteristics of the binding conformation of the antibody to be designed CDRH3-antigen target epitope.

[0107] S20: Based on the sequence length of the CDRH3 region of the antibody to be designed, select multiple antibody sequences from the antibody sequence library whose CDRH3 region sequence length is the same as that of the antibody to be designed, and use them as candidate antibody sequences.

[0108] For example, if the CDRH3 region of the antibody to be designed has a sequence length of 10, then antibody sequences with a CDRH3 region length of 10 are selected from the antibody sequence library as candidate antibody sequences.

[0109] S21: Compare the number of different amino acids contained in the CDRH3 region of the candidate antibody sequence with the number of different amino acids contained in the CDRH3 region of the antibody to be designed, and select the candidate antibody sequence with the fewest different amino acids as the template sequence.

[0110] Since the candidate antibody sequence is only the same as the CDRH3 region of the antibody to be designed, but the amino acid sequences contained in the CDRH3 region are not completely the same, it is necessary to select the candidate antibody sequence with the fewest different amino acids as the template sequence so that the CDRH3 region of the template sequence is basically the same as the CDRH3 region of the antibody to be designed in terms of sequence length and amino acid sequence.

[0111] S30: Replace the CDRH3 region of the template sequence with the CDRH3 region of the antibody to be designed, and use it as a candidate antibody sequence.

[0112] In this embodiment, the CDRH3 region of the antibody to be designed is the CDR3 region of the heavy chain sequence. The antibody sequence library contains both heavy chain (IGHV) and light chain (IGKV / IGLV) sequences. Therefore, during replacement, the CDR3 region of the heavy chain sequence in the template sequence is replaced with the CDRH3 region of the antibody to be designed. It should be noted that the light chain sequence of the antibody to be designed can be set according to business needs or experience; this application does not limit this.

[0113] The candidate antibody sequence obtained by the replacement in step S30 has both a complete sequence and retains the CDRH3 region sequence of the antibody to be designed.

[0114] S40: Based on the candidate antibody sequence, predict the conformation of the candidate antibody.

[0115] The embodiments of this application may employ algorithms such as AlphaFold and Immubuilder, which can predict the three-dimensional structure of antibodies, to predict the conformation of candidate antibodies based on the candidate antibody sequence. This application does not limit this to any particular method.

[0116] S50: Based on a preset algorithm, the conformation of the CDRH3 region of the candidate antibody is optimized by adding constraints to obtain multiple optimized antibody conformations. The constraint settings can make the conformations of the CDRH3 region of the multiple optimized antibodies meet the binding characteristics of the CDRH3-antigen target epitope binding conformation of the antibody to be designed and the conformational diversity characteristics of the CDRH3 region of the candidate antibody.

[0117] In some embodiments, the constraints of this application may include at least one of condition one, condition two, condition three, and condition four, wherein condition one is: constraining the dihedral angles of corresponding residue sites on the CDRH3 region of the candidate antibody based on the residue sites on the CDRH3 region of the antibody to be designed that interact with the target epitope of the antigen; condition two is: constraining the distance of hydrogen bonds in conserved regions within the CDRH3 region of the candidate antibody; condition three is: based on the conformational clustering results of antibody CDRH3 regions in existing databases, according to the sequence length of the CDRH3 region of the antibody to be designed, obtaining the corresponding conformationally conserved residue sites in each clustering result, and constraining the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody after excluding the residue sites constrained in condition one; condition four is: constraining the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody according to a set method after excluding the residue sites constrained in conditions one and three.

[0118] Among these conditions, condition one can constrain the phi and psi angles of the interacting residues on the CDRH3 region of the candidate antibody, making the structure of the optimized antibody CDRH3 similar to the structure of the designed antibody CDRH3 in the binding conformation of the designed antibody CDRH3-antigen epitope; condition two can constrain the hydrogen bond distance between the conserved regions FWRH3 and FWRH4 at both ends of the candidate antibody CDRH3, preventing excessive structural perturbations or even unreasonable changes during the simulation process, and maintaining the stability of the antibody during the simulation calculation; condition three is to make the conformation of the candidate antibody CDRH3 conform to the existing conformation and to diversify the candidate antibody conformation; and condition four is to further generate more diverse CDRH3 conformations.

[0119] For example, the CDRH3 region of a candidate antibody has a sequence length of 10, containing 10 numbered amino acids (corresponding to IMGT site numbers 105, 106, 107, 108, 109, 113, 114, 115, 116, 117). In the binding conformation of the candidate antibody CDRH3 to the target epitope generated by the HERN model, amino acids at positions 4, 5, and 6 of the candidate antibody CDRH3 can bind to the target epitope. Therefore, amino acids at positions 4, 5, and 6 (numbered 108, 109, 113) are also interacting amino acids in the CDRH3 region of the candidate antibody. Condition 1 constrains the psi and phi dihedral angles of amino acids 108, 109, and 113 in the CDRH3 region of the candidate antibody.

[0120] The CDRH3 conformation clustering results for the CDRH3 region of the antibody to be designed, with a sequence length of 10, may result in 3 classes. For each class, the mean and variance of the psi and phi angles of the 10 amino acids in the CDRH3 region of the antibody to be designed are calculated. The sites with lower variance are conserved sites, generally the first and last few amino acids, such as the 1st, 2nd, 3rd, and 9th and 10th (amino acids at positions 4, 5, and 6 are excluded). Condition 3 constrains the psi and phi dihedral angles of the amino acids at the corresponding sites of the candidate antibody sequence (corresponding to IMGT site numbers 105, 106, 107, 116, and 117). Condition 4 constrains the psi and phi dihedral angles of the remaining sites in the CDRH3 region of the candidate antibody, such as the 7th and 8th amino acids (corresponding to IMGT site numbers 114 and 115), according to the set method.

[0121] In condition four, the configuration method is randomly generated to produce as diverse a configuration as possible. Examples of this configuration method are given below:

[0122] For example, after conditions one and three, there are still 6 residue sites that have not been subjected to dihedral constraints. Condition four can randomly select a number of sites from these 6 sites (e.g., IMGT sites numbered 106, 107, 108, 114, 115, and 116) to generate dihedral constraints. If the number of selected sites is less than 5, 2 sets of constraints are generated; if it is greater than or equal to 5, 4 sets of constraints are generated. Specifically, if sites 106, 108, and 115 are randomly selected and dihedral constraints are applied to these three sites to form one set of constraints, then another set of constraints needs to be randomly generated. If sites 106, 108, and 115 are randomly selected again, since they are repeated from the first set, a new set of constraints needs to be selected. If sites 106, 108, and 116 are selected again and are not repeated from the first set, then dihedral constraints are applied to these three sites to form another set of constraints, thus forming two sets of constraints.

[0123] Additionally, it should be noted that conditions 1, 3, and 4 have a priority order: condition 1 > condition 3 > condition 4. If the CDRH3 sequence is very short and conditions 1 and 3 already cover all amino acids, then the constraint of condition 4 will not be added.

[0124] The preset algorithm in the embodiments of this application can be a protein molecule simulation algorithm, and this application does not limit it.

[0125] S60: Based on the binding conformation of the target epitope of the antibody to be designed CDRH3-antigen, adjust the positions of multiple optimized antibodies until the position of the CDRH3 region of the optimized antibody basically overlaps with the position of the CDRH3 region of the antibody to be designed, and obtain multiple optimized antibody-antigen complex conformations and sequences.

[0126] In this embodiment, based on the existing binding conformation of the antibody-antigen target epitope CDRH3 to be designed, the positions of multiple optimized antibodies can be adjusted by means of, for example, alignment (e.g., the spatial coordinates of the optimized antibody in a computer interface) until the position of the CDRH3 region of the optimized antibody basically overlaps with the position of the CDRH3 region of the antibody to be designed in the binding conformation of the antibody-antigen target epitope CDRH3 to be designed (e.g., the spatial coordinate difference between the CDRH3 region of the optimized antibody and the CDRH3 region of the antibody to be designed is the smallest in a computer interface). This yields the relative position data of the optimized antibody and the antigen that can bind to the antigen target epitope. Based on the relative position data, the conformation and sequence of multiple optimized antibody-antigen complexes are obtained.

[0127] S70: Based on multiple optimized antibody-antigen complex conformations and optimized antibody-antigen interactions, screen for optimized antibody-antigen complex conformations and sequences that meet the second preset screening criteria.

[0128] In some embodiments, the second preset screening condition in the embodiments of this application may be: the conformations of the optimized antibody and antigen in the optimized antibody-antigen complex conformation do not conflict, and the interaction (e.g., the number of interacting residues) between the optimized antibody and antigen in regions other than CDRH3 in the optimized antibody-antigen complex conformation is not less than a threshold.

[0129] In this embodiment of the application, the binding conformation of the antibody and antigen in the antibody-antigen complex is optimized, as well as the interaction between the antibody and antigen is optimized. Complexes with conflicting binding conformations that prevent binding and have weak interactions are removed. The remaining complexes have optimized antibody and antigen conformations that do not conflict and can bind, and have strong interactions with a high probability and degree of binding.

[0130] S80: The binding interface of the screened optimized antibody-antigen complex is structurally and sequentially optimized to obtain the structure and sequence of multiple target antibodies that can bind to the target epitopes of the antigen.

[0131] This application embodiment optimizes the binding interface of the antibody-antigen complex conformation that has passed screening, thereby structurally enhancing the interaction of the antibody-antigen binding interface and increasing the success rate of antibody design.

[0132] The embodiments of this application can optimize the binding interface of the remaining antibody-antigen complex conformation using interface optimization algorithms such as protein molecule simulation algorithms and ProteinMPNN, and this application does not limit this.

[0133] After step S80, antibodies that can truly bind to the target epitope of the antigen can be screened from the target antibodies through phage surface display.

[0134] Compared to the local conformation of antibody CDRH3 binding to the target epitope in related models, the embodiments of this application can obtain the complete antibody-antigen conformation of antibody CDRH3 binding to the target epitope, and simulate the diversity of the antibody CDRH3 region, generating diverse antibody conformations. The complete antibody-antigen conformation not only determines the interaction between all regions of the antibody, including CDRH3 (e.g., CDRL1, CDRL2, CDRL3, CDRH1, CDRH2, FWRH, FWRL), and all epitope regions of the antigen, including the target epitope, but also determines the interaction between non-CDRH3 regions and other epitopes of the antigen. This eliminates antibody-antigen complex conformations that cannot bind, as well as complex conformations with weak antibody-antigen interactions, providing more accurate and higher-success-rate antibody data for subsequent experimental screening.

[0135] Corresponding to the aforementioned application function implementation method embodiments, this application also provides an antibody design device, electronic device, and corresponding embodiments.

[0136] Figure 4 This is a schematic diagram of the antibody design device shown in the embodiments of this application.

[0137] See Figure 4 This application provides an antibody design device, including an acquisition module 410, a first screening module 420, a replacement module 430, a prediction module 440, a first optimization module 450, a docking module 460, a second screening module 470, and a second optimization module 480.

[0138] The acquisition module 410 is used to acquire the sequence of the CDRH3 region of the antibody to be designed and the binding conformation of the CDRH3 region of the antibody to be designed with the antigen target epitope; wherein, the acquisition module 410 is used to generate a number of CDRH3 regions of the antibody to be designed corresponding to the sequence length according to the conformation of the antigen and the target epitope, as well as the preset sequence length of the CDRH3 region of the antibody to be designed; select a preset number of sequences from all the CDRH3 regions of the antibody to be designed; and generate multiple binding conformations of the CDRH3 region of the antibody to be designed with the antigen target epitope according to each selected CDRH3 region sequence of the antibody to be designed.

[0139] The screening module 420 is used to select antibody sequences that meet the first preset screening conditions from the antibody sequence library based on the sequence of the CDRH3 region of the antibody to be designed, and use them as template sequences. Specifically, the screening module 420 is used to select multiple antibody sequences from the antibody sequence library whose CDRH3 region sequence length is the same as that of the CDRH3 region of the antibody to be designed, and use them as candidate antibody sequences. The screening module 420 compares the number of different amino acids contained in the CDRH3 region sequence of the candidate antibody sequences with the number of different amino acids contained in the CDRH3 region sequence of the antibody to be designed, and selects the candidate antibody sequence with the fewest different amino acids as the template sequence.

[0140] The replacement module 430 is used to replace the CDRH3 region of the template sequence with the CDRH3 region of the antibody to be designed, and use it as a candidate antibody sequence.

[0141] The prediction module 440 is used to predict the conformation of the candidate antibody based on the candidate antibody sequence.

[0142] The first optimization module 450 is used to optimize the conformation of the CDRH3 region of the candidate antibody by adding constraints based on a preset algorithm, thereby obtaining the conformations of multiple optimized antibodies. The constraint settings can make the conformations of the CDRH3 region of the multiple optimized antibodies meet the binding characteristics of the CDRH3-antigen target epitope binding conformation of the antibody to be designed and the conformational diversity characteristics of the CDRH3 region of the candidate antibody. The constraints in this application embodiment may include at least one of condition one, condition two, condition three, and condition four. Condition one is: constraining the dihedral angles of corresponding residue sites on the CDRH3 region of the candidate antibody based on the residue sites on the CDRH3 region of the antibody to be designed that interact with the target epitope of the antigen; condition two is: constraining the distance of hydrogen bonds in conserved regions within the CDRH3 region of the candidate antibody; condition three is: based on the conformational clustering results of antibody CDRH3 regions in existing databases, obtaining the corresponding conformationally conserved residue sites in each clustering result according to the sequence length of the CDRH3 region of the antibody to be designed, and constraining the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody after excluding the residue sites already constrained in condition one; condition four is: constraining the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody according to a set method after excluding the residue sites already constrained in conditions one and three.

[0143] The docking module 460 is used to dock the CDRH3 regions of multiple optimized antibodies with the target epitopes of antigens to obtain multiple optimized antibody-antigen complex conformations and sequences. The docking module 460 is used to adjust the positions of multiple optimized antibodies according to the binding conformation of the CDRH3 region of the antibody to be designed with the target epitope of the antigen, until the positions of the CDRH3 regions of the optimized antibodies and the CDRH3 regions of the antibody to be designed are basically superimposed, thus obtaining multiple optimized antibody-antigen complex conformations and sequences.

[0144] The second screening module 470 is used to screen optimized antibody-antigen complex conformations and sequences that meet the second preset screening conditions based on multiple optimized antibody-antigen complex conformations and optimized antibody-antigen interactions. The second preset screening conditions in this embodiment may be: the optimized antibody and antigen conformations do not conflict in the optimized antibody-antigen complex conformations, and the interaction between the optimized antibody and antigen in the region outside CDRH3 in the optimized antibody-antigen complex conformations is not lower than a threshold.

[0145] The second optimization module 480 is used to optimize the structure and sequence of the binding interface of the optimized antibody-antigen complex conformation that has passed the screening, so as to obtain the structure and sequence of multiple target antibodies that can bind to the target epitopes of the antigen.

[0146] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated further here.

[0147] Figure 5 This is a schematic diagram of the structure of an electronic device shown in an embodiment of this application.

[0148] See Figure 5 The electronic device 500 includes a memory 510 and a processor 520.

[0149] The processor 520 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0150] Memory 510 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. ROM may store static data or instructions required by the processor 520 or other modules of the computer. Permanent storage devices may be read-write storage devices. Permanent storage devices may be non-volatile storage devices that retain stored instructions and data even when the computer is powered off. In some embodiments, permanent storage devices use mass storage devices (e.g., magnetic or optical disks, flash memory) as permanent storage devices. In other embodiments, permanent storage devices may be removable storage devices (e.g., floppy disks, optical drives). System memory may be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. System memory may store some or all of the instructions and data required by the processor during operation. Furthermore, memory 510 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, memory 510 may include a removable storage device that is readable and / or writable, such as a laser disc (CD), a read-only digital multifunction optical disc (e.g., DVD-ROM, dual-layer DVD-ROM), a read-only Blu-ray disc, an ultra-high-density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not contain carrier waves or transient electronic signals transmitted wirelessly or via wired connections.

[0151] The memory 510 stores executable code, which, when processed by the processor 520, can cause the processor 520 to execute part or all of the methods described above.

[0152] Furthermore, the method according to this application can also be implemented as a computer program or computer program product, which includes computer program code instructions for performing some or all of the steps in the method described above.

[0153] Alternatively, this application may be implemented as a computer-readable storage medium (or a non-transitory machine-readable storage medium or a machine-readable storage medium) storing executable code (or computer program or computer instruction code) thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the steps of the methods described above according to this application.

[0154] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for designing antibodies, characterized in that, include: Obtain the sequence of the CDRH3 region of the antibody to be designed and the binding conformation of the antibody to the target epitope CDRH3-antigen; Based on the sequence of the CDRH3 region of the antibody to be designed, an antibody sequence that meets the first preset screening condition is selected from the antibody sequence library as a template sequence; The CDRH3 region of the template sequence is replaced with the CDRH3 region of the antibody to be designed, and this is used as a candidate antibody sequence. Based on the candidate antibody sequence, the conformation of the candidate antibody is predicted; Based on a preset algorithm, the conformation of the CDRH3 region of the candidate antibody is optimized by adding constraints, resulting in multiple optimized antibody conformations. The constraints ensure that the conformations of the CDRH3 regions of the multiple optimized antibodies satisfy both the binding characteristics of the CDRH3-antigen target epitope binding conformation of the antibody to be designed and the conformational diversity characteristics of the CDRH3 region of the candidate antibody. The constraints include at least one of conditions one, two, three, and four. Condition one is: based on the residue sites on the CDRH3 region of the antibody to be designed that interact with the antigen target epitope, the corresponding residue sites on the CDRH3 region of the candidate antibody are constrained. The conditions are as follows: Condition 2: Constrain the distance of hydrogen bonds within conserved regions of the CDRH3 region of the candidate antibody; Condition 3: Based on the conformational clustering results of antibody CDRH3 regions in the existing database, according to the sequence length of the CDRH3 region of the antibody to be designed, obtain the corresponding conformationally conserved residue sites in each clustering result, exclude the residue sites already constrained in Condition 1, and constrain the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody; Condition 4: After excluding the residue sites already constrained in Conditions 1 and 3, constrain the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody according to a set method. Based on the binding conformation of the antibody CDRH3-antigen target epitope to be designed, the CDRH3 regions of multiple optimized antibodies are docked with the antigen target epitope to obtain multiple optimized antibody-antigen complex conformations and sequences. Based on the multiple optimized antibody-antigen complex conformations and the interactions between the optimized antibody and the antigen, optimized antibody-antigen complex conformations and sequences that meet the second preset screening conditions are screened. The binding interface of the screened optimized antibody-antigen complexes was structurally and sequentially optimized to obtain the structures and sequences of multiple target antibodies that can bind to antigen target epitopes.

2. The method according to claim 1, characterized in that, The process of obtaining the sequence of the CDRH3 region of the antibody to be designed and the binding conformation of the CDRH3-antigen target epitope of the antibody to be designed includes: Based on the conformation of the antigen and the target epitope, and the preset sequence length of the CDRH3 region of the antibody to be designed, generate a number of sequences of the CDRH3 region of the antibody to be designed corresponding to the sequence length; Select a predetermined number of sequences from all the CDRH3 regions of the antibodies to be designed; Based on the selected sequence of each CDRH3 region of the antibody to be designed, multiple binding conformations of the antibody to be designed CDRH3-antigen target epitopes are generated.

3. The method according to claim 1, characterized in that, The step of selecting antibody sequences that meet the first preset screening criteria from the antibody sequence library as template sequences based on the sequence of the CDRH3 region of the antibody to be designed includes: Based on the sequence length of the CDRH3 region of the antibody to be designed, multiple antibody sequences with the same CDRH3 region sequence length as the antibody to be designed are selected from the antibody sequence library and used as candidate antibody sequences. Compare the number of different amino acids contained in the CDRH3 region of the candidate antibody sequence with the number of different amino acids contained in the CDRH3 region of the antibody to be designed, and screen out the candidate antibody sequence with the fewest different amino acids as the template sequence; and / or, The second preset filtering conditions include: In the optimized antibody-antigen complex conformation, the conformations of the optimized antibody and the antigen do not conflict, and in the optimized antibody-antigen complex conformation, the interaction between the optimized antibody and the antigen in regions other than CDRH3 is not lower than a threshold.

4. The method according to claim 1, characterized in that, The method further includes: Based on the binding conformation of the antibody CDRH3-antigen target epitope to be designed, the binding characteristics of the binding conformation of the antibody CDRH3-antigen target epitope to be designed are extracted. The binding characteristics include the interaction residue pairs between the antibody CDRH3 region to be designed and the antigen target epitope, as well as the dihedral angles of the corresponding residue sites on the antibody CDRH3 region to be designed.

5. The method according to claim 4, characterized in that, The step of extracting the binding characteristics of the CDRH3-antigen target epitope of the antibody to be designed, based on its binding conformation, includes: Based on the binding conformation of the antibody CDRH3 to the target epitope of the antigen, calculate the distance between the CDRH3 region of the antibody to be designed and the residue pair on the target epitope of the antigen. The sites where residue pairs with a distance less than a preset value are located are taken as interacting residue pairs, and the dihedral angles of the interacting residue pairs and the residue sites located on the CDRH3 region of the antibody to be designed are taken as the binding characteristics of the binding conformation of the antibody to be designed CDRH3-antigen target epitope.

6. The method according to claim 1, characterized in that, The step involves docking the CDRH3 regions of multiple optimized antibodies with the target epitope of the antigen according to the binding conformation of the CDRH3-antigen target epitope of the antibody to be designed, thereby obtaining multiple optimized antibody-antigen complex conformations and sequences, including: Based on the binding conformation of the target epitope of the antibody to be designed CDRH3-antigen, the positions of multiple optimized antibodies are adjusted until the position of the CDRH3 region of the optimized antibody basically overlaps with the position of the CDRH3 region of the antibody to be designed, thereby obtaining multiple optimized antibody-antigen complex conformations and sequences.

7. An antibody design device, characterized in that, include: The acquisition module is used to acquire the sequence of the CDRH3 region of the antibody to be designed and the binding conformation of the CDRH3-antigen target epitope of the antibody to be designed; The first screening module is used to select antibody sequences that meet the first preset screening conditions from the antibody sequence library as template sequences based on the sequence of the CDRH3 region of the antibody to be designed. The replacement module is used to replace the CDRH3 region of the template sequence with the CDRH3 region of the antibody to be designed, and use it as a candidate antibody sequence. A prediction module is used to predict the conformation of the candidate antibody based on the candidate antibody sequence; The first optimization module is used to optimize the conformation of the CDRH3 region of the candidate antibody by adding constraints based on a preset algorithm, thereby obtaining multiple optimized antibody conformations. The constraints are set to ensure that the conformations of the CDRH3 regions of the multiple optimized antibodies satisfy the binding characteristics of the CDRH3-antigen target epitope binding conformation of the antibody to be designed and the conformational diversity characteristics of the CDRH3 region of the candidate antibody. The constraints include at least one of conditions one, two, three, and four, wherein condition one is: constraining the CDRH3 region of the candidate antibody based on the residue sites on the CDRH3 region of the antibody to be designed that interact with the antigen target epitope. The conditions are as follows: Condition 1: Constrain the dihedral angles of the corresponding residue sites; Condition 2: Constrain the distance of hydrogen bonds within the conserved regions of the CDRH3 region of the candidate antibody; Condition 3: Based on the conformational clustering results of antibody CDRH3 regions in the existing database, according to the sequence length of the CDRH3 region of the antibody to be designed, obtain the corresponding conformationally conserved residue sites in each clustering result, exclude the residue sites already constrained in Condition 1, and constrain the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody; Condition 4: After excluding the residue sites already constrained in Conditions 1 and 3, constrain the dihedral angles of the remaining residue sites on the CDRH3 region of the candidate antibody according to the set method. The docking module is used to dock the CDRH3 region of multiple optimized antibodies with the target epitope of the antigen according to the binding conformation of the CDRH3-antigen target epitope of the antibody to be designed, so as to obtain multiple optimized antibody-antigen complex conformations and sequences. The second screening module is used to screen optimized antibody-antigen complex conformations and sequences that meet the second preset screening conditions based on the multiple optimized antibody-antigen complex conformations and the interaction between the optimized antibody and the antigen. The second optimization module is used to optimize the structure and sequence of the binding interface of the screened optimized antibody-antigen complex to obtain the structure and sequence of multiple target antibodies that can bind to the target epitopes of the antigen.

8. An electronic device, characterized in that, include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any one of claims 1-6.

9. A computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the method as described in any one of claims 1-6.

Images