System and method for generating protein sequences for therapeutic agent in consideration of human safety

Abstract

The present invention relates to a system and a method for generating protein sequence information. The system comprises: a memory for storing one or more instructions; and at least one processor for executing the one or more instructions stored in the memory, wherein an operation performed by the one or more instructions comprises the steps of: acquiring reference protein sequence information; repeatedly adding noise to the reference protein sequence information to generate protein sequence information to which noise has been added; and generating output protein sequence information by removing noise from the protein sequence information to which noise has been added, wherein the step of generating the output protein sequence information comprises repeatedly performing all or some of the steps of: removing noise from input protein sequence information according to at least one protein structure guidance specified by a user to generate protein sequence information from which noise has been removed; and adding noise to the protein sequence information from which noise has been removed according to at least one protein sequence guidance specified by the user, thereby generating protein sequence information to which noise has been added.

Description

System and method for generating protein sequences for therapeutic purposes considering human safety

[0001] The present disclosure relates to a system and method for generating a protein amino acid sequence having desired characteristics. More specifically, the present disclosure relates to a system and method for generating an amino acid sequence for a protein that has excellent disease-treating effects and is safe for the human body, making it suitable for use as a disease treatment.

[0002] With the recent advancement of artificial intelligence (AI), efforts are being made to leverage AI to shorten research and development periods and increase efficiency in the field of protein therapeutic drug development. Attempts are ongoing to train AI models with accumulated data on protein amino acid sequences, structures, properties, and functions to generate protein amino acid sequences predicted to have binding properties to desired targets. However, developing protein therapeutics requires consideration of both target binding properties and human safety. However, research on AI models that generate therapeutic proteins that fully consider human safety remains inadequate.

[0003] Therapeutic protein drugs are larger than traditional chemically synthesized drugs, and thus can interact with human immune cells and cause unexpected side effects. These unexpected side effects can be fatal, even fatal. Therefore, pharmaceutical companies developing therapeutic protein drugs must predict potential side effects and design highly safe therapeutic protein drugs. However, accurately predicting these side effects before administering therapeutic protein drugs to humans remains challenging, requiring significant time and expense to assess human safety.

[0004] Therefore, an artificial intelligence system and method are needed to produce therapeutic proteins that are both safe for the human body and have excellent therapeutic effects, while also fully considering the side effects that may occur when administered to the human body.

[0005] The present disclosure seeks to provide a system and method capable of generating protein expressions that reflect protein interaction partner information.

[0006] Through this disclosure, it is intended to provide a system and method for predicting the physical properties, structure, binding force, interaction, etc. of a protein from protein data or designing an amino acid sequence of a protein having desired properties, and to provide a system and method with higher prediction reliability compared to systems previously utilized.

[0007] One embodiment of the present disclosure can provide a protein sequence information generation system.

[0008] One embodiment of the present disclosure may provide a system including a memory storing one or more instructions; and at least one processor executing the one or more instructions stored in the memory.

[0009] The actions performed by one or more of the above commands are

[0010] Step of obtaining reference protein sequence information,

[0011] A step of repeatedly adding noise to reference protein sequence information to generate protein sequence information with added noise, and

[0012] A step of generating output protein sequence information from which noise has been removed from protein sequence information to which noise has been added,

[0013] Here, the above creation step is

[0014] A step of removing noise from input protein sequence information according to one or more user-specified protein structure guidances to generate denoised protein sequence information, and

[0015] It may include a step of repeatedly performing all or part of the step of adding noise to the denoised protein sequence information according to one or more protein sequence guidances specified by the user to generate protein sequence information with added noise.

[0016] In one embodiment of the present disclosure, it may be characterized by generating protein sequence information having a property and protein motif that binds to a target protein specified by a user.

[0017] In one embodiment of the present disclosure, the target protein may be characterized by being at least one protein selected from among proteins associated with one or more of the occurrence, treatment, prevention, and alleviation of a human disease.

[0018] In one embodiment of the present disclosure, it may be characterized by generating protein sequence information of all or part of an antibody or a binding fragment thereof.

[0019] In one embodiment of the present disclosure, it may be characterized by generating protein sequence information including amino acid sequence information of a complementary binding site of an antibody or a binding fragment thereof.

[0020] In one embodiment of the present disclosure, the structural guidance may be at least one selected from the group consisting of binding affinity for the target protein, immunogenicity for B cells, and off-target binding affinity.

[0021] In one embodiment of the present disclosure, the sequence guidance may be at least one selected from the group consisting of B cell immunogenicity, and helper T cell immunogenicity.

[0022] One embodiment of the present disclosure may provide a method for generating protein sequence information performed by at least one processor.

[0023] The above method comprises the steps of obtaining reference amino acid sequence information,

[0024] A step of repeatedly adding noise to reference amino acid sequence information to generate protein sequence information with added noise, and

[0025] A step of generating output protein sequence information from which noise has been removed from protein sequence information to which noise has been added,

[0026] Here, the above creation step is

[0027] A step of removing noise from input protein sequence information according to one or more user-specified protein structure guidances to generate denoised protein sequence information, and

[0028] A method for generating protein sequence information, comprising the step of repeatedly performing all or part of the step of adding noise to the denoised protein sequence information according to one or more protein sequence guidances specified by a user to generate protein sequence information with added noise.

[0029] In one embodiment of the present disclosure, the method may be characterized by generating protein sequence information having a property of binding to a target protein specified by a user.

[0030] In one embodiment of the present disclosure, the target protein may be characterized by being at least one protein selected from among proteins associated with one or more of the occurrence, treatment, prevention, and alleviation of a human disease.

[0031] In one embodiment of the present disclosure, the method may be characterized by generating protein sequence information of all or part of an antibody or a binding fragment thereof.

[0032] In one embodiment of the present disclosure, the method may be characterized by generating protein sequence information including amino acid sequence information of a complementary binding site of an antibody or a binding fragment thereof.

[0033] In one embodiment of the present disclosure, the structural guidance may be at least one selected from the group consisting of binding affinity for the target protein, immunogenicity for B cells, and off-target binding affinity.

[0034] In one embodiment of the present disclosure, the sequence guidance may be at least one selected from the group consisting of B cell immunogenicity, and helper T cell immunogenicity.

[0035] One embodiment of the present disclosure can provide a program stored on a computer-readable recording medium to execute the above method on a computer.

[0036] According to one embodiment of the present disclosure, a user can obtain protein sequence information having desired characteristics.

[0037] According to one embodiment of the present disclosure, amino acid sequence information of a therapeutic protein having excellent disease treatment effect and excellent human safety can be obtained, wherein the therapeutic protein has a property of binding to a target protein designated by a user and has a protein motif designated by the user.

[0038] By utilizing the amino acid sequence of a therapeutic protein produced according to one embodiment of the present disclosure, the cost and time required for new drug development can be significantly reduced.

[0039] FIG. 1 is a flowchart illustrating a system and method for generating output protein sequence information by repeating steps of adding noise to reference protein sequence information and then removing noise according to one embodiment of the present disclosure.

[0040] FIG. 2 is a flowchart illustrating an iterative process of removing noise from protein sequence information to which noise has been added and then adding noise again according to one embodiment of the present disclosure, and illustrating steps in which user-specified structural guidance and sequence guidance are reflected in the iterative process.

[0041] FIG. 3 is a block diagram illustrating a device for generating protein sequence information from reference protein sequence information according to one embodiment of the present disclosure.

[0042] To clarify the technical idea of ​​the present disclosure, embodiments of the present disclosure will be described in detail with reference to the attached drawings. In describing the present disclosure, if a detailed description of a related known function or component is determined to unnecessarily obscure the gist of the present disclosure, the detailed description will be omitted. In the drawings, components having substantially the same function or configuration are given the same reference numbers and symbols as possible even if they are shown in different drawings. For convenience of explanation, devices and methods are described together when necessary. Each operation of the present disclosure does not necessarily have to be performed in the order described and may be performed in parallel, selectively, or individually.

[0043] The terms used in the embodiments of this disclosure have been selected from widely used, current terms, taking into account the functions of the present disclosure. However, these terms may vary depending on the intentions of those skilled in the art, precedents, the emergence of new technologies, etc. Furthermore, in certain cases, the applicant may arbitrarily select terms, and in such cases, their meanings will be described in detail in the description of the relevant embodiments. Therefore, the terms used in this specification should not be defined simply as names of terms, but rather based on their meanings and the overall content of the present disclosure.

[0044] Throughout this disclosure, singular expressions may include plural expressions unless the context clearly dictates otherwise. Terms such as "comprise" or "have" should be understood to indicate the presence of a feature, number, step, operation, component, part, or combination thereof, but do not preclude the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof. In other words, when it is said throughout this disclosure that a part "comprises" a certain component, unless specifically stated otherwise, this does not mean that other components may be included, but rather that other components may be excluded.

[0045] Expressions such as "at least one" modify the entire list of elements, not individual elements of the list. For example, "at least one of A, B, and C" and "at least one of A, B, or C" refer to only A, only B, only C, both A and B, both B and C, both A and C, all of A, B, and C, or any combination thereof.

[0046] In addition, terms such as “...unit”, “...module”, etc. described in the present disclosure mean a unit that processes at least one function or operation, which may be implemented as hardware or software, or a combination of hardware and software.

[0047] Throughout this disclosure, when a part is said to be "connected" to another part, this includes not only cases where the parts are "directly connected," but also cases where the parts are "electrically connected" with other elements intervening. Furthermore, when a part is said to "include" a component, this does not exclude other components, but rather includes other components, unless otherwise specifically stated.

[0048] The expression “configured to” as used throughout this disclosure can be used interchangeably with, for example, “suitable for,” “having the capacity to,” “designed to,” “adapted to,” “made to,” or “capable of.” The term “configured to” does not necessarily mean something that is “specifically designed to” in terms of hardware. Instead, in some contexts, the expression “a system configured to” can mean that the system, together with other devices or components, is “capable of.” For example, the phrase “a processor configured (or set) to perform A, B, and C” may mean a dedicated processor (e.g., an embedded processor) for performing those operations, or a generic-purpose processor (e.g., a CPU or application processor) that can perform those operations by executing one or more software programs stored in memory.

[0049] The artificial intelligence-related functions according to the present disclosure are operated via a processor and memory. The processor may be comprised of one or more processors. In this case, one or more processors may be a general-purpose processor such as a CPU, an AP, a Digital Signal Processor (DSP), a graphics-only processor such as a GPU or a Vision Processing Unit (VPU), or an artificial intelligence-only processor such as an NPU. One or more processors control the processing of input data according to predefined operating rules or artificial intelligence models stored in memory. Alternatively, if one or more processors are artificial intelligence-only processors, the artificial intelligence-only processor may be designed with a hardware structure specialized for processing a specific artificial intelligence model.

[0050] The predefined operation rules or artificial intelligence models are characterized by being created through learning. Here, "created through learning" means that the predefined operation rules or artificial intelligence models are trained using learning data by a learning algorithm, thereby creating predefined operation rules or artificial intelligence models that are set to achieve a desired purpose. This learning may be performed on the device itself on which the artificial intelligence according to the present disclosure is performed, or may be performed through a separate server and / or system.

[0051] Throughout the present disclosure, devices may include, but are not limited to, servers, smartphones, tablet PCs, PCs, TVs, smart TVs, mobile phones, personal digital assistants (PDAs), speakers, laptops, media players, microservers, e-book object recognition devices, digital broadcasting object recognition devices, kiosks, MP3 players, digital cameras, robot vacuum cleaners, home appliances, other mobile or non-mobile computing devices, watches, glasses, hair bands, and rings having communication and data processing capabilities.

[0052] The present disclosure relates to a system and method for generating protein sequence information using a noise-based diffusion model. The diffusion model comprises a forward diffusion step that progressively adds noise to data, which disrupts the data, and a reverse denoising step that transforms the noisy data into denoised data. The diffusion model is trained using a deep learning model by parameterizing the reverse transformation step, and the trained deep learning model performs a reverse wandering transformation to generate denoised data from the noisy data.

[0053] FIG. 1 illustrates a system and method for generating output protein sequence information by repeating the steps of adding noise to reference protein sequence information using a diffusion model and then removing the noise, according to one embodiment of the present disclosure.

[0054] Obtaining reference protein sequence information (101)

[0055] Reference protein sequence information refers to information regarding the amino acid sequence of a reference protein. According to one embodiment of the present disclosure, the reference protein sequence information uses information extracted from the amino acid sequence of a previously known protein. According to one embodiment of the present disclosure, the protein amino acid sequence that serves as the basis for the reference protein sequence information may be selected from amino acid sequences of proteins known to bind to a user-specified target protein and to have a user-specified protein motif.

[0056] Generating protein sequence information with added noise (102)

[0057] According to one embodiment of the present disclosure, protein sequence information with added noise can be generated through a diffusion step of gradually adding noise that disrupts data to the acquired reference protein sequence information.

[0058] According to one embodiment of the present disclosure, one or more noise models selected from the group consisting of Gaussian noise, salt-and-pepper noise, Poisson noise, binomial noise, and speckle noise may be used, and a system or method may be constructed by appropriately selecting a noise model according to an application of the diffusion model, or by combining one or more noise models, and a user may also specify the type of noise model used in the system and method of the present disclosure. In one embodiment of the present disclosure, a Gaussian noise model is preferably used, but is not limited thereto.

[0059] According to one embodiment of the present disclosure, the diffusion step includes a step of progressively adding noise over multiple stages. According to one embodiment of the present disclosure, noise can be added to reference protein sequence information, and noise can be repeatedly added to the noise-added information to ultimately generate protein sequence information with added noise.

[0060] Generating protein sequence information with noise removed (103)

[0061] According to one embodiment of the present disclosure, protein sequence information predicted to have desired characteristics by a user can be generated by removing noise from protein sequence information to which noise has been added using an artificial neural network.

[0062] According to one embodiment of the present disclosure, the artificial neural network can remove noise from protein sequence information to which noise has been added to generate protein sequence information from which noise has been removed, and a protein having a protein amino acid sequence extracted from the protein sequence information from which noise has been removed can be predicted to have a protein structure exhibiting characteristics desired by the user.

[0063] According to one embodiment of the present disclosure, the step of removing noise from protein sequence information to which noise has been added and generating denoised protein sequence information may be performed using an artificial neural network model trained by minimizing a loss function. According to one embodiment of the present disclosure, the step of removing noise may comprise multiple steps of progressively removing noise.

[0064] Generating protein sequence information with added noise (104)

[0065] By repeating the steps of adding noise again to the protein sequence information from which noise has been removed (104, 212) and then removing noise again (103, 211) several times, highly reliable protein sequence information can be obtained.

[0066] According to one embodiment of the present disclosure, the same noise model as the noise model used in the diffusion step (102) for adding noise to the reference protein sequence information is used, and a Gaussian noise model is used as the noise model, but is not limited thereto.

[0067] Generate denoised output protein sequence information (105)

[0068] By repeatedly performing the steps of removing noise from protein sequence information with added noise (103, 211) and adding noise again to the protein sequence information with removed noise (104, 212), highly reliable output protein sequence information can be obtained.

[0069] Figure 2 is a flowchart specifically showing the denoising step and the step in which the structural guidance and sequence guidance are applied in the flowchart shown in Figure 1.

[0070] Step to remove noise in the denoising stage (211)

[0071] According to one embodiment of the present disclosure, the step of removing noise from protein sequence information to which noise has been added to generate protein sequence information from which noise has been removed can be performed using an artificial neural network model learned by minimizing a loss function.

[0072] According to one embodiment of the present disclosure, the step of removing noise (211) uses an artificial neural network trained to generate protein sequence information predicted to have characteristics according to structural guidance.

[0073] According to one embodiment of the present disclosure, a property associated with the structure of a protein may be used as a structural guidance. This may be at least one selected from the group consisting of binding affinity for the target protein, immunogenicity for B cells, and off-target binding affinity.

[0074] Binding affinity to a target protein refers to the binding properties of a therapeutic protein to the target protein. The stronger the binding affinity, the greater the amount of target protein binding can be achieved with a smaller amount of therapeutic protein. The stronger the binding affinity to the target protein, the more likely it is that the therapeutic protein will be effective in treating disease.

[0075] B-cell immunogenicity refers to the property of a therapeutic protein to elicit an immune response in the body, mediated by B cells. If administration of a therapeutic protein triggers an excessive immune response in B cells, potentially fatal side effects can occur. Therefore, a therapeutic protein with lower B-cell immunogenicity is expected to be safer in the human body.

[0076] Off-target binding affinity refers to the ability of a therapeutic protein to bind to a biopolymer (such as a protein) in the human body other than the target protein. When a therapeutic protein is administered intravenously, binding to a biopolymer other than the target protein increases the likelihood of adverse effects. Therefore, a lower off-target binding affinity indicates a therapeutic protein with superior human safety.

[0077] According to one embodiment of the present disclosure, the artificial neural network can remove noise from sequence information data to which noise has been added, thereby generating amino acid sequence information of a protein having characteristics set by a user as a structural guidance.

[0078] Step 2: Adding noise in the denoising stage (212)

[0079] By repeating the step of adding noise again to the protein sequence information from which noise has been removed (212) and then removing noise again (211) several times, highly reliable protein sequence information can be obtained.

[0080] According to one embodiment of the present disclosure, the same noise model as the noise model used in the diffusion step (102) for adding noise to the reference protein sequence information may be used, and preferably, a Gaussian noise model may be used as the noise model, but is not limited thereto.

[0081] According to one embodiment of the present disclosure, a property associated with the sequence of a protein may be used as a sequence guide. This may be one or more selected from the group consisting of immunogenicity against B cells and immunogenicity against helper T cells.

[0082] Helper T cells are immune cells that can be activated by fragmented foreign proteins (therapeutic proteins). The immunogenicity of these fragmented foreign proteins to helper T cells can be determined by the amino acid sequence of the fragmented foreign protein. If a therapeutic protein contains an amino acid sequence capable of activating helper T cells, it may have high immunogenicity to helper T cells. In this case, the therapeutic protein may induce an excessive and unwanted immune response, potentially causing fatal side effects. Therefore, the lower the immunogenicity to helper T cells, the safer the therapeutic protein is predicted to be in the human body.

[0083] B cells are activated by binding to a therapeutic protein structure, and both the structure and protein sequence of the therapeutic protein can influence whether or not B cells are activated. Therefore, according to one embodiment of the present disclosure, immunogenicity for B cells can also be used as a sequence guide.

[0084] According to one embodiment of the present disclosure, amino acid sequence information of a protein having characteristics set by a user as a sequence guidance can be generated by repeating the steps of adding noise to sequence information data from which noise has been removed and removing noise again.

[0085] FIG. 3 is a block diagram of a protein expression learning device according to one embodiment of the present disclosure.

[0086] Referring to FIG. 3, the protein expression learning device (300) may include a transceiver (310), a memory (320), a database (330), and a processor (340). However, not all of the components illustrated in FIG. 3 are essential components of the protein expression learning device (300). The protein expression learning device (300) may be implemented with more components than the components illustrated in FIG. 3, or may be implemented with fewer components than the components illustrated in FIG. 3. In addition, the transceiver (310), the memory (320), and the processor (340) may be implemented in the form of a single chip.

[0087] In one embodiment, the transceiver (310) can communicate with a terminal or other electronic device connected wired or wirelessly to the protein expression learning device (300). For example, the transceiver (310) can obtain protein amino acid sequence information, protein interaction data, or protein expressions generated using artificial neural networks from the other electronic device.

[0088] Various types of data, such as programs and files, such as applications, can be installed and stored in the memory (320). The processor (340) can access and use data stored in the memory (320), or store new data in the memory (320). In addition, the memory (320) can store one or more instructions. The processor (340) can execute one or more instructions stored in the memory.

[0089] The processor (340) controls the overall operation of the protein expression learning device (300) and may include at least one processor, such as a CPU or a GPU. The processor (340) may control other components included in the protein expression learning device (300) to perform operations for operating the protein expression learning device (300). For example, the processor (340) may acquire protein data, acquire protein expressions using the neural network, obtain contrast loss from the protein expressions, and modify one or more values ​​of one or more parameters of one or more encoder neural networks based on the contrast loss.

[0090] The database (330) can store various learning data for training a learning model. Furthermore, the database (330) can store protein amino acid sequence information, protein interaction data, protein structure information, simulation result information, and the like. In various embodiments, the database can also store output data generated by the learning model. While FIG. 3 illustrates the protein expression learning device (300) as including the database (330), the database (330) may be provided externally to the device. In this case, the database (330) can be connected to the protein expression learning device (300) via wired or wireless connections.

[0091] Additionally, the learning model may be implemented outside the protein expression learning device (300) (e.g., cloud-based) or may be included within the protein expression learning device (300).

[0092] An embodiment of the present disclosure may also be implemented in the form of a recording medium containing computer-executable instructions, such as program modules, executed by a computer. Computer-readable media may be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Furthermore, computer-readable media may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Communication media typically contains computer-readable instructions, data structures, or program modules, and includes any information delivery media.

[0093] The above description of the present disclosure is provided for illustrative purposes only, and those skilled in the art will readily appreciate that the present disclosure can be readily modified into other specific forms without altering the technical spirit or essential characteristics of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, components described as being single may be implemented in a distributed manner, and similarly, components described as being distributed may be implemented in a combined manner.

[0094] The scope of the present disclosure is indicated by the claims described below rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present disclosure.

Claims

1. In a protein sequence information generation system, memory for storing one or more instructions; and At least one processor for executing one or more instructions stored in the memory, The actions performed by one or more of the above commands are Step of obtaining reference protein sequence information, A step of repeatedly adding noise to reference protein sequence information to generate protein sequence information with added noise, and A step of generating output protein sequence information from which noise has been removed from protein sequence information to which noise has been added, Here, the above creation step is A step of removing noise from input protein sequence information according to one or more user-specified protein structure guidances to generate denoised protein sequence information, and A protein sequence information generation system comprising a step of repeatedly performing all or part of a step of adding noise to the denoised protein sequence information according to one or more protein sequence guidances specified by a user to generate protein sequence information with added noise.

2. A protein sequence information generation system characterized in that it generates protein sequence information having a property of binding to a target protein designated by a user and a protein motif in accordance with the first paragraph.

3. A protein sequence information generation system according to claim 2, wherein the target protein is at least one protein selected from among proteins associated with at least one of the occurrence, treatment, prevention, and alleviation of a human disease.

4. A protein sequence information generation system characterized in that it generates protein sequence information of all or part of an antibody or a binding fragment thereof in the third paragraph.

5. A protein sequence information generation system characterized in that it generates protein sequence information including amino acid sequence information of a complementary binding site of an antibody or a binding fragment thereof in the fourth paragraph.

6. A protein sequence information generation system in the second paragraph, wherein the structural guidance is at least one selected from the group consisting of binding affinity for the target protein, immunogenicity for B cells, and off-target binding affinity.

7. A protein sequence information generation system in the second paragraph, wherein the sequence guidance is at least one selected from the group consisting of B cell immunogenicity and helper T cell immunogenicity.

8. A method for generating protein sequence information performed by at least one processor, The above method comprises the steps of obtaining reference amino acid sequence information, A step of repeatedly adding noise to reference amino acid sequence information to generate protein sequence information with added noise, and A step of generating output protein sequence information from which noise has been removed from protein sequence information to which noise has been added, Here, the above creation step is A step of removing noise from input protein sequence information according to one or more user-specified protein structure guidances to generate denoised protein sequence information, and A method for generating protein sequence information, comprising the step of repeatedly performing all or part of the step of adding noise to the denoised protein sequence information according to one or more protein sequence guidances specified by a user to generate protein sequence information with added noise.

9. A protein sequence information generation method characterized in that, in paragraph 8, protein sequence information having a property of binding to a target protein designated by a user is generated.

10. A method for generating protein sequence information, characterized in that in claim 9, the target protein is at least one protein selected from among proteins associated with at least one of the occurrence, treatment, prevention, and alleviation of a human disease.

11. A method for generating protein sequence information, characterized in that it generates protein sequence information of all or part of an antibody or a binding fragment thereof in accordance with claim 10.

12. A method for generating protein sequence information, characterized in that it generates protein sequence information including amino acid sequence information of a complementary binding site of an antibody or a binding fragment thereof in accordance with claim 11.

13. A method for generating protein sequence information in claim 9, wherein the structural guidance is at least one selected from the group consisting of binding affinity for the target protein, immunogenicity for B cells, and off-target binding affinity.

14. A method for generating protein sequence information in claim 9, wherein the sequence guidance is at least one selected from the group consisting of B cell immunogenicity and helper T cell immunogenicity.

15. A program stored on a computer-readable recording medium that causes a computer to execute the method of any one of clauses 8 to 14.

Images