Information processing system, information processing method, information processing program, and molecular compound production method

Abstract

An information processing system is provided with at least one processor. The at least one processor performs the following processes: for each of a plurality of first molecules, acquiring training data indicating a plurality of feature quantities associated with the first molecule; acquiring, for each of a plurality of second molecules, test data indicative of a plurality of characteristic quantities associated with the second molecule; for each of the plurality of feature quantities, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being a probability distribution of the feature quantities in the training data, and the second probability distribution being a probability distribution of the feature quantities in the test data; and selecting, based on the index for each of the plurality of feature quantities, one or more of the plurality of feature quantities as one or more input feature quantities to be used as input parameters of a prediction model that predicts a feature value of a molecule based on machine learning.

Description

(19) State Intellectual Property Office (12) Invention Patent Application (10) Application Publication Number (43) Application Publication Date (21) Application Number 202380104703.6 (22) Application Date 2023.12.21 (85) PCT International Application Entering National Phase Date 2026.06.10 (86) PCT International Application Application Data PCT / JP2023 / 045988 2023.12.21 (87) PCT International Application Publication Data WO2025 / 134327 JA 2025.06.26 (71) Applicant Chugai Pharmaceutical Co., Ltd. Address Japan (72) Inventors Yoshimura Oshirai and Eiji Honrei and Takizawa Hideki (74) Patent Agency Zhongke Patent & Trademark Agency Co., Ltd. 11021 Patent Attorney Zhu Dan (51) Int.Cl. G06N 20 / 00 (2006.01) (54) Invention Title: Information Processing System, Information Processing Method, Information Processing Program, and Method for Manufacturing Molecular Compounds (57) Abstract: An information processing system is provided with at least one processor. The at least one processor performs the following processing: for each of a plurality of first molecules, acquiring training data indicating a plurality of feature quantities associated with the first molecule; for each of a plurality of second molecules, acquiring test data indicating a plurality of feature quantities associated with the second molecule; for each of the plurality of feature quantities, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being the probability distribution of the feature quantity in the training data and the second probability distribution being the probability distribution of the feature quantity in the test data; and based on the index for each of the plurality of feature quantities, selecting one or more of the plurality of feature quantities as one or more input feature quantities to be used as input parameters for a predictive model based on machine learning to predict the property values ​​of molecules.Claims (3 pages), Description (24 pages), Drawings (11 pages), CN 122374769 A, 2026.07.10, CN 1 22 37 47 69 A. 1. An information processing system comprising at least one processor, wherein the at least one processor performs the following processing: for each of a plurality of first molecules, acquiring training data representing a plurality of feature values ​​associated with the first molecule; for each of a plurality of second molecules, acquiring test data representing a plurality of feature values ​​associated with the second molecule; for each of the plurality of feature values, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being a probability distribution in the training data and the second probability distribution being a probability distribution in the test data; and based on the index of each of the plurality of feature values, selecting one or more of the plurality of feature values ​​as one or more input feature values ​​to be used as input parameters for a predictive model for predicting property values ​​of molecules based on machine learning. 2. The information processing system of claim 1, wherein the at least one processor performs the following processing: for each of the plurality of feature values, calculating an inter-distribution distance as the index, the inter-distribution distance being the distance between the first probability distribution and the second probability distribution; and selecting one or more input feature values ​​from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values. 3. The information processing system of claim 2, wherein the at least one processor selects one or more feature values ​​whose inter-distribution distance is less than a predetermined reference as the one or more input feature values. 4. The information processing system of claim 2 or 3, wherein the at least one processor calculates a Wasserstein distance as the inter-distribution distance. 5. The information processing system of any one of claims 2 to 4, wherein the at least one processor performs the following processing: excluding a portion of the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values, and selecting the remaining one or more feature values ​​as the one or more input feature values.6. The information processing system of claim 5, wherein the at least one processor performs the following processes: repeated evaluation processing, wherein in the evaluation processing, the number of feature values ​​excluded from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values ​​is changed, while generating a provisional prediction model using the training data through the machine learning and calculating an evaluation metric for the provisional prediction model based on the remaining one or more feature values; determining the number of feature values ​​excluded from the plurality of feature values ​​as an exclusion quantity based on the plurality of evaluation metrics; and excluding the exclusion quantity of feature values ​​from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values, and selecting the one or more input feature values. 7. The information processing system according to claim 6, wherein the plurality of feature values ​​are N feature values, where N is a natural number greater than 2, and the at least one processor performs the following processing: (Claim 1 / 3 page 2 CN 122374769 A) Setting the number of feature values ​​to be excluded from the N feature values ​​to n, where n is a natural number; excluding n feature values ​​from the plurality of feature values ​​based on the inter-distribution distance for each of the N feature values; performing an evaluation process, wherein the generation of a first provisional prediction model using the training data through the machine learning and the calculation of an evaluation index for the first provisional prediction model are performed based on (N-n) feature values; further excluding n feature values ​​from the (N-n) feature values ​​based on the inter-distribution distance for each of the (N-n) feature values; and performing an evaluation process, wherein the generation of a second provisional prediction model using the training data through the machine learning and the calculation of an evaluation index for the second provisional prediction model are performed based on (N-2n) feature values. 8. The information processing system according to claim 6 or 7, wherein the at least one processor determines the number of times the highest evaluation index is obtained as the exclusion quantity, or determines the number of times the lowest evaluation index is obtained as the exclusion quantity.9. The information processing system according to any one of claims 2 to 4, wherein the at least one processor performs the following processes: repeated evaluation processing, wherein in the evaluation processing, the number of feature values ​​selected from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values, while generating a provisional prediction model using the training data through the machine learning based on the selected one or more feature values ​​and calculating an evaluation metric for the provisional prediction model; determining the number of feature values ​​selected from the plurality of feature values ​​as a selection quantity based on the plurality of evaluation metrics; and selecting the selected number of feature values ​​as the one or more input feature values ​​based on the inter-distribution distance for each of the plurality of feature values. 10. The information processing system according to any one of claims 6 to 9, wherein the at least one processor performs the evaluation processing by cross-validation. 11. The information processing system according to any one of claims 1 to 10, wherein the at least one processor generates the prediction model by performing the machine learning using one or more input feature values ​​of the training data. 12. The information processing system of claim 11, wherein the at least one processor performs the following processes: acquiring one or more input feature values ​​for a molecule of interest, the characteristic value of the molecule of interest being unknown; and outputting the characteristic value of the molecule of interest, the characteristic value being obtained by inputting the acquired one or more input feature values ​​into the generated prediction model. 13. The information processing system of claim 11, wherein the at least one processor performs the following processes: acquiring one or more input feature values ​​for each of a plurality of molecules of interest, the characteristic value of each of the plurality of molecules of interest being unknown; inputting the one or more input feature values ​​acquired for each of the plurality of molecules of interest into the generated prediction model, and obtaining the characteristic value of the molecule of interest from the prediction model; selecting at least one molecule of interest from the plurality of molecules of interest based on the characteristic value of each of the plurality of molecules of interest; and outputting information about the at least one selected molecule of interest. 14. The information processing system according to any one of claims 1 to 13, wherein each of the first molecule and the second molecule is selected from nucleic acids, peptides, cyclic peptides, proteins, antibodies and low molecular weight compounds.15. An information processing method, executed by an information processing system including at least one processor, the information processing method comprising the steps of: acquiring training data representing a plurality of feature values ​​associated with each of a plurality of first molecules; acquiring test data representing a plurality of feature values ​​associated with each of a plurality of second molecules; calculating an index based on a first probability distribution and a second probability distribution for each of the plurality of feature values, the first probability distribution being a probability distribution in the training data and the second probability distribution being a probability distribution in the test data; and selecting one or more of the plurality of feature values ​​as one or more input feature values ​​based on the index of each of the plurality of feature values ​​for use as input parameters of a predictive model for predicting molecule property values ​​based on machine learning. 16. An information processing program that enables a computer to perform the following steps: for each of a plurality of first molecules, acquiring training data representing a plurality of feature values ​​associated with the first molecule; for each of a plurality of second molecules, acquiring test data representing a plurality of feature values ​​associated with the second molecule; for each of the plurality of feature values, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being a probability distribution in the training data and the second probability distribution being a probability distribution in the test data; and selecting one or more of the plurality of feature values ​​as one or more input feature values ​​based on the index of each of the plurality of feature values ​​to be used as input parameters for a predictive model for predicting property values ​​of molecules based on machine learning. 17. A method for manufacturing a molecular compound, the method comprising: a generation step of generating a molecular compound having a molecular sequence of the at least one molecule of interest based on information output from an information processing system of claim 13. Claims 3 / 3 Page 4 CN 122374769 A Information processing system, information processing method, information processing program and method for manufacturing molecular compounds Technical Field

[0001] This disclosure relates to an information processing system, an information processing method, an information processing program and a method for generating molecular compounds. Background Art

[0002] Patent Document 1 discloses a method for identifying the amino acid sequence of an antibody with affinity for an antigen. The method includes the steps of: querying a machine learning engine for a proposed amino acid sequence of an antibody with high affinity for an antigen; and obtaining the proposed amino acid sequence from the machine learning engine.

[0003] Non-Patent Document 1 discloses a technique for selecting feature values ​​in training data and test data.In this technique, when the score of the classification between the training data and the test data obtained by the adversarial classifier is higher than a predetermined threshold, the feature values ​​that are highly important in the classification are excluded, and the adversarial classifier is retrained. When the score is lower than the threshold, the remaining feature values ​​are used to train the prediction model.

[0004] [Citation List]

[0005] [Patent Documents]

[0006] [Patent Document 1] International Publication No. WO 2018 / 132752

[0007] [Non-Patent Documents]

[0008] [Non-Patent Document 1] Pan, Jing et al., “Adversarial validation approach to concept drift problem in user targeting automation systems at uber.” arXiv: 2004.03045 (2020). Summary of the Invention

[0009] [Technical Problem]

[0010] It is desirable to improve the accuracy of machine learning models used to predict the properties of molecules.

[0011] [Solution to the Problem]

[0012] The information processing system according to an aspect of this disclosure includes at least one processor. The at least one processor performs the following processing: for each of a plurality of first molecules, acquiring training data representing a plurality of feature values ​​associated with the first molecule; for each of a plurality of second molecules, acquiring test data representing a plurality of feature values ​​associated with the second molecule; for each of the plurality of feature values, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being a probability distribution in the training data and the second probability distribution being a probability distribution in the test data; and based on the index for each of the plurality of feature values, selecting one or more of the plurality of feature values ​​as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecule property values ​​based on machine learning.

[0013] The information processing method according to an aspect of the present disclosure is performed by an information processing system including at least one processor.The information processing method includes the following steps: for each of a plurality of first molecules, acquiring training data representing a plurality of feature values ​​associated with the first molecule; for each of a plurality of second molecules, acquiring test data representing a plurality of feature values ​​associated with the second molecule; for each of the plurality of feature values, calculating an index based on a first probability distribution and a second probability distribution, wherein the first probability distribution is the probability distribution in the training data and the second probability distribution is the probability distribution in the test data; and based on the index for each of the plurality of feature values, selecting one or more of the plurality of feature values ​​as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecule property values ​​based on machine learning.

[0014] The information processing procedure according to aspects of this disclosure enables a computer to perform the following steps: for each of a plurality of first molecules, acquiring training data representing a plurality of feature values ​​associated with the first molecule; for each of a plurality of second molecules, acquiring test data representing a plurality of feature values ​​associated with the second molecule; for each of the plurality of feature values, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being a probability distribution in the training data and the second probability distribution being a probability distribution in the test data; and based on the index for each of the plurality of feature values, selecting one or more of the plurality of feature values ​​as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecule property values ​​based on machine learning.

[0015] In these aspects, the feature value to be used as an input parameter for the predictive model based on machine learning is selected as an input feature value based on the trend of each feature value in both the training data and the test data. By selecting feature values ​​as described above, the accuracy of the machine learning model (predictive model) used to predict molecule properties can be improved.

[0016] [Beneficial Effects of the Invention]

[0017] According to aspects of this disclosure, the accuracy of machine learning models used to predict the properties of molecules can be improved. Brief Description of the Drawings

[0018] [Figure 1] Figure 1 shows a diagram illustrating the selection of input feature values.

[0019] [Figure 2] Figure 2 shows a diagram illustrating an example of the functional configuration of an information processing system.

[0020] [Figure 3] Figure 3 shows a diagram illustrating an example of the hardware configuration of a computer acting as an information processing system.

[0021] [Figure 4] Figure 4 shows a diagram illustrating an example of training data.

[0022] [Figure 5] Figure 5 shows a flowchart illustrating an example of a process performed by the information processing system.

[0023] [Figure 6] Figure 6 shows a flowchart illustrating an example of a process for selecting input feature values.

[0024] [Figure 7] Figure 7 shows a diagram for demonstrating cross-validation.

[0025] [Figure 8] Figure 8 shows a flowchart demonstrating an example of cross-validation.

[0026] [Figure 9] Figure 9 shows a flowchart demonstrating another example of the process for selecting input feature values.

[0027] [Figure 10] Figure 10 shows a diagram demonstrating a first example of validation.

[0028] [Figure 11] Figure 11 shows a diagram demonstrating a second example of validation. Detailed Description

[0029] Hereinafter, various embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same or equivalent elements have the same reference numerals, and redundant descriptions thereof are omitted.

[0030] [System Summary]

[0031] The information processing system according to the present disclosure is a computer system that selects one or more feature values ​​of a molecule, which are used as input parameters for a predictive model for predicting the property values ​​of the molecule based on machine learning. In the present disclosure, the feature values ​​used as input parameters are referred to as "input feature values". The information processing system eliminates some feature values ​​from a plurality of candidate feature values ​​for input feature values, and then selects the remaining one or more feature values ​​as one or more input feature values. The number of feature values ​​is also called the number of dimensions. The process for selecting one or more input feature values ​​from a plurality of feature values ​​can also be described as a process for reducing the number of dimensions of feature values.

[0032] In an example, the information processing system uses the selected one or more input feature values ​​to perform machine learning, thereby generating a relevant predictive model. This process corresponds to the learning phase of machine learning. In an example, the information processing system inputs one or more input feature values ​​of a molecule whose feature values ​​are unknown into the generated predictive model, thereby predicting the feature values ​​of the molecule. In this disclosure, the molecule whose feature values ​​are predicted by the information processing system is also called a "molecule of interest". The information processing system may perform screening, wherein the feature value of each of a plurality of molecules of interest is predicted, and at least one molecule of interest is selected from the plurality of molecules of interest based on the predicted feature values. The prediction of feature values ​​corresponds to the prediction phase or operation phase of machine learning. In the learning phase and the prediction phase, the information processing system generates and uses a predictive model with the selected one or more input feature values.

[0033] Machine learning is a technique used to autonomously discover patterns or rules based on given information through iterative learning. Predictive models generated by machine learning are machine learning models built using algorithms and data structures, and are also called learned models. In this example, the predictive model is built using neural networks such as convolutional neural networks (CNNs).

[0034] In this disclosure, feature values ​​are numerical values ​​representing the quantified characteristics of a molecule. In this example, multiple feature values ​​are set for each molecule.Each feature value can represent a characteristic of the relationship between multiple constituent units forming a molecule, a sequence of multiple constituent units, or a characteristic of a specific constituent unit.

[0035] In this disclosure, a feature value is a numerical value representing a quantitative characteristic of a molecule. For example, a feature value can represent various characteristics related to a molecule's binding ability, affinity, pharmacological activity, physical properties, kinetics, or safety.

[0036] The information processing system selects input feature values. A prediction model is generated using training data on the molecule through machine learning. Based on the prediction model, predicted values ​​of the molecule's characteristics shown by the test data are calculated. In this disclosure, the generated prediction model is also referred to as a "characteristic prediction model".

[0037] In this disclosure, each of the molecules shown by the training data is called a "first molecule". Each of the molecules shown by the test data is called a "second molecule". The first molecule and the second molecule can be morphologically identical or different. As an example, the first molecule is an antibody, and the second molecule is an antibody. As another example, the first molecule is a cyclic peptide, and the second molecule is a cyclic peptide. For example, when using the energy of the interaction between a ligand and a protein as a feature value to predict affinity, the first molecule can be a small molecule, and the second molecule is a cyclic peptide.

[0038] The first molecule group includes a plurality of first molecules. For each of the plurality of first molecules, the training data includes a plurality of feature values ​​associated with the first molecule. The second molecule group includes a plurality of second molecules. For each of the plurality of second molecules, the test data includes a plurality of feature values ​​associated with the second molecule. The plurality of first molecules in the first molecule group can be morphologically identical or different. The plurality of second molecules in the second molecule group can be morphologically identical or different.

[0039] As an example, when the molecule is an antibody, training data representing the antibody sequence is used. In the case of molecules having the sequence information described above, the first molecule group can be rewritten as a sequence group for training, and the second molecule group can be rewritten as a sequence group for prediction.

[0040] For each of the plurality of first molecules, the training data includes a plurality of feature values ​​of the first molecule and at least one feature value of the first molecule. For each of the plurality of second molecules, the test data includes a plurality of feature values ​​of the second molecule. If a relatively large number of feature values ​​exhibit a changing trend between the training and test data, the accuracy of the predictive model will deteriorate. In this disclosure, the information processing system selects one or more input feature values ​​such that feature values ​​exhibiting a changing trend between the training and test data do not affect the machine learning and predictive model. In one example, the information processing system selects feature values ​​with relatively small trend differences between the training and test data. In another example, the information processing system excludes feature values ​​with significantly different trends between the training and test data.Instruction Manual 3 / 24 Page 7 CN 122374769 A

[0041] In an example, the information processing system calculates the inter-distribution distance as a selection criterion for each of a plurality of feature values. The inter-distribution distance is the distance between a first probability distribution and a second probability distribution, where the first probability distribution is the probability distribution in the training data and the second probability distribution is the probability distribution in the test data. The probability distribution of the feature value is a distribution that shows the probability of each value assumed by the feature value. Since the probability distribution represents the trend of the feature value, it can be said that feature values ​​with relatively large differences between the first probability distribution and the second probability distribution have different trends between the training data and the test data. The selection criterion is an indicator used to select input feature values.

[0042] Figure 1 shows a graph used to illustrate the selection of input feature values. In this example, the information processing system calculates the inter-distribution distance as a selection criterion for each of Q feature values. The inter-distribution distance is the distance between the first probability distribution 210 in the training data and the second probability distribution 220 in the test data. The information processing system excludes feature values ​​(e.g., feature values ​​F1, F3, F4) that have relatively large differences between the first probability distribution 210 and the second probability distribution 220, and then selects the remaining feature values ​​(e.g., feature values ​​F2, F5, FQ) as input feature values. Each of the selected input feature values ​​is a feature value that has relatively small differences between the first probability distribution 210 and the second probability distribution 220 and has a relatively consistent trend. Therefore, the prediction model generated by machine learning based on the input feature values ​​can predict the characteristic values ​​of the molecules of interest.

[0043] [System Configuration]

[0044] Figure 2 shows a diagram illustrating the functional configuration of the information processing system 10 according to an example. In this example, the information processing system 10 accesses a database 20 for storing training data for machine learning. The database 20 may be provided in a different computer system than the information processing system 10, or it may be a component of the information processing system 10. In this example, the information processing system 10 accesses the database 20 via a communication network such as the Internet or an intranet.

[0045] The information processing system 10 includes a feature value selector 11, a learner 12, and a predictor 13. The feature value selector 11 is a functional module that selects one or more input feature values ​​from a plurality of feature values. The learner 12 is a functional module that generates a prediction model 30 based on the selected one or more input feature values ​​through machine learning. The predictor 13 is a functional module that performs predictions related to molecules of interest using the generated prediction model 30.

[0046] FIG3 shows a diagram illustrating an example of the hardware configuration of the computer 100 that serves as the information processing system 10.For example, computer 100 includes processor 101, main memory 102, secondary memory 103, communication controller 104, input device 105, and output device 106. Processor 101 executes operating system and applications. Main memory 102 includes, for example, ROM and RAM. Secondary memory 103 includes hard disk or flash memory and typically stores a larger amount of data than main memory 102. Communication controller 104 includes, for example, network interface card or wireless communication module. Input device 105 includes, for example, keyboard, mouse, and touch panel. Output device 106 includes, for example, monitor and speaker.

[0047] The functional modules of information processing system 10 are implemented by information processing program 110 pre-stored in secondary memory 103. Specifically, the functional modules are implemented by allowing information processing program 110 to be read on processor 101 or main memory 102, wherein processor 101 executes information processing program 110. Processor 101 causes communication controller 104, input device 105, or output device 106 to operate according to information processing program 110 and to read and write data from main memory 102 or auxiliary memory 103. Data, predictive models, or databases required for processing may be stored in main memory 102 or auxiliary memory 103.

[0048] Information processing program 110 may be recorded in a non-transitory computer-readable storage medium (such as CD-ROM, DVD-ROM, or semiconductor memory) and provided. Alternatively, information processing program 110 may be provided via a communication network as a data signal superimposed on a carrier wave.

[0049] Information processing system 10 may include one computer 100, or may include multiple computers 100. When multiple computers 100 are used, the computers 100 are connected via a communication network (such as the Internet or an intranet) to logically construct an information processing system 10.

[0050] Figure 4 illustrates an example of training data stored in database 20. The training data includes multiple data records corresponding to multiple molecules. In the example, each data record includes a molecule ID, which uniquely identifies each molecule as a data item, multiple feature values ​​of the molecule, and property values ​​of the molecule. When the molecule is a protein, the multiple feature values ​​used as training data can be obtained, for example, by evaluating a protein embedding method (TAPE). In the example of Figure 4, database 20 stores R data records, each representing Q feature values ​​(i.e., Q-dimensional feature values).The number of dimensions Q of the feature values ​​can be less than 10, or it can be on the order of tens, hundreds, thousands, or tens of thousands. The information processing system 10 excludes some feature values ​​from the Q feature values, and then selects one or more remaining feature values ​​as one or more input feature values. In Example 4, physical property values ​​are represented as molecular property values, but property values ​​are not limited to physical property values.

[0051] [System Operation]

[0052] Figure 5 shows a flowchart illustrating an example of a process performed by the information processing system 10 (as process flow S1). Process flow S1 is an example of an information processing method according to the present disclosure. In step S10, the feature value selector 11 selects one or more input feature values ​​from a plurality of feature values ​​represented by training data for use in machine learning. In the example, the feature value selector 11 calculates a plurality of evaluation metrics, some of which feature values ​​are selected or excluded, wherein the evaluation metrics are calculated while changing the feature values ​​to be selected or excluded. The feature value selector 11 selects a set of feature values ​​as input feature values ​​when the evaluation metrics meet predetermined conditions. In step S20, learner 12 generates prediction model 30 by performing machine learning based on one or more selected input feature values. In step S30, predictor 13 performs prediction using prediction model 30. Step S10 corresponds to the preprocessing of the learning phase, step S20 corresponds to the learning phase, and step S30 corresponds to the prediction phase. One or more input feature values ​​selected in step S10 are used in steps S20 and S30.

[0053] (Selection of input feature values)

[0054] FIG6 shows a flowchart illustrating in detail an example of the process for selecting input feature values ​​(i.e., an example of step S10).

[0055] In step S11, feature value selector 11 obtains training data and test data from database 20. For each of the plurality of first molecules, the training data includes a plurality of feature values ​​associated with the first molecule. For each of the plurality of second molecules, the test data includes a plurality of feature values ​​associated with the second molecule.

[0056] In step S12, based on the first probability distribution in the training data and the second probability distribution in the test data, the feature value selector 11 calculates a selection metric for each of a plurality of feature values. This selection metric is used to select the input feature value. In this example, the feature value selector 11 calculates the inter-distribution distance as the selection metric for each feature value. This inter-distribution distance is the distance between the first probability distribution and the second probability distribution. The inter-distribution distance indicates the degree of difference between the first probability distribution and the second probability distribution. The feature value selector 11 may use the integral probability metric (IPM) as the inter-distribution distance.As an example of IPM, the feature selector 11 can calculate the Wasserstein distance, maximum mean difference (MMD), or Dudley metric as the distance between distributions.

[0057] The feature selector 11 can calculate a first probability distribution and a second probability distribution for calculating the selection metric. The feature value is Fi, the number of data records in the training data is j, and the number of data records in the test data is k. The feature selector 11 can calculate the probability distribution of j values ​​for the feature value Fi as the first probability distribution and calculate the probability distribution of k values ​​as the second probability distribution.

[0058] In step S13, the feature selector 11 initializes the exclusion quantity n of the feature values. The exclusion quantity n is the number of feature values ​​excluded from the multiple feature values ​​based on the selection metric of each of the multiple feature values, as described on page 5 / 24 of the specification, CN 122374769 A. The exclusion quantity n can also be described as the numerical reduction of the dimension of the feature value. The feature selector 11 changes the exclusion quantity n in any of the subsequent processes. When the number of exclusions n increases, the initial value of the numerical reduction n can be 0, 1 or more. When the number of exclusions n decreases, the initial value of the numerical reduction n can be {(total number of eigenvalues) - 1} or a smaller number.

[0059] In step S14, the eigenvalue selector 11 excludes n eigenvalues ​​from a plurality of eigenvalues ​​based on the selection criterion for each eigenvalue. When using inter-distribution distances such as Wasserstein distance, the eigenvalue selector 11 excludes n eigenvalues ​​in descending order of inter-distribution distances. That is, the eigenvalue selector 11 excludes n eigenvalues ​​in which the distance between the first probability distribution and the second probability distribution is relatively large.

[0060] In step S15, the eigenvalue selector 11 generates and evaluates a provisional prediction model based on one or more eigenvalues ​​remaining after the exclusion process. In this disclosure, the generation and evaluation of the provisional prediction model are also referred to as "evaluation processing". The provisional prediction model is not the prediction model 30 generated by learner 12 and used by predictor 13, but a machine learning model (learned model) temporarily used to select one or more input feature values. Feature selector 11 may use cross-validation or retention methods as the method for evaluating the processing.

[0061] Cross-validation will be described as an example of step S15. In cross-validation, feature selector 11 divides the training data into multiple groups, selects one of the multiple groups as validation data, and selects the remaining groups as narrowly defined training data. Each of the independent groups is also referred to as a "fold". The combination of validation data and narrowly defined training data is also referred to as a "split".Feature selector 11 generates a provisional prediction model using narrowly defined training data and evaluates the provisional prediction model using validation data. Feature selector 11 performs the generation and evaluation of the provisional prediction model while changing the group (fold) used as validation data and calculates the evaluation metric of the provisional prediction model for each split. Feature selector 11 obtains statistics of multiple evaluation metrics obtained from multiple splits as evaluation metrics for one cross-validation.

[0062] Figure 7 shows a diagram for illustrating cross-validation. In the example shown in the figure, feature selector 11 divides the training data into five groups. In the first split, feature selector 11 selects "fold 1" as validation data and selects the remaining four groups as narrowly defined training data. Feature selector 11 generates a provisional prediction model through machine learning using the narrowly defined training data and calculates the evaluation metric E1 of the provisional prediction model using the validation data. Feature selector 11 generates and evaluates provisional prediction models for the second to fifth segments while changing the validation data, and calculates four evaluation metrics E2, E3, E4, and E5 for the segments. Feature selector 11 determines the statistics of the five evaluation metrics E1 to E5 as the final evaluation metrics for one cross-validation. Examples of statistics include the mean and median, but are not limited to these.

[0063] Figure 8 shows a flowchart illustrating an example of cross-validation. In step S151, feature selector 11 sets the cross-validation segments. Feature selector 11 divides the training data obtained from database 20 into multiple groups, selects one of these groups as validation data, and selects the remaining groups as narrowly defined training data.

[0064] In step S152, feature selector 11 generates a provisional prediction model by performing machine learning based on one or more feature values ​​remaining after the exclusion process. Feature selector 11 generates a provisional prediction model by performing machine learning (supervised learning) using narrowly defined training data. Feature selector 11 does not use the excluded n features and feeds the remaining one or more features as input parameters (e.g., input vectors) into the machine learning model. In this example, feature selector 11 updates the parameter set in the machine learning model by performing backpropagation (the backpropagation method) based on the error between the predicted values ​​computed by the machine learning model and the correct solution (label). Learner 12 obtains a provisional predictive model by repeating this process until a given termination condition is met. The termination condition may be the processing of all data records of the narrowly defined training data.Instruction manual 6 / 24 pages 10 CN 122374769 A

[0065] In step S153, the feature selector 11 calculates an evaluation metric based on the remaining one or more feature values. The evaluation metric is a value that indicates how accurately the provisional prediction model can calculate the feature value. The feature selector 11 does not use the excluded n feature values ​​and inputs the remaining one or more feature values ​​as input parameters (e.g., input vectors) into the provisional prediction model for each data record of the validation data. The provisional prediction model calculates the feature value based on the input parameters for each data record. Hereinafter, the feature value calculated by the provisional prediction model is also referred to as the predicted feature value.

[0066] The feature selector 11 calculates the evaluation metric of the provisional prediction model based on the predicted feature value and the correct solution (label) for each data record of the validation data. The correct solution (label) in the process is the feature value in the training data used as the validation data. For example, the feature selector 11 may calculate the mean squared error between the predicted feature value and the correct solution, or calculate a correlation coefficient indicating the degree of correlation between the predicted feature value and the correct solution.

[0067] As shown in step S154, when there are still unexecuted segmentations (No in step S154), the process returns to step S151. In the repeated step S151, the feature selector 11 changes the group (fold) used as validation data and sets the next segment. In the repeated step S152, the feature selector 11 generates a provisional prediction model by performing machine learning (supervised learning) on ​​the training data with a narrow definition based on the remaining one or more feature values. In the repeated step S153, the feature selector 11 calculates the evaluation metric of the provisional prediction model based on the remaining one or more feature values.

[0068] When all segmentations have been processed (Yes in step S154), the process continues to step S155. In step S155, the feature selector 11 calculates the statistics of the evaluation metric for each segment as the final evaluation metric for cross-validation. For example, the feature selector 11 calculates the average mean squared error or the average correlation coefficient as the final evaluation metric.

[0069] Similarly, referring to FIG6, in step S16, the feature value selector 11 determines whether to change the exclusion quantity n. The change in the exclusion quantity n can be an increase or a decrease.

[0070] In an example, the feature value selector 11 determines whether to increase the exclusion quantity n based on the number of feature values ​​excluded by the exclusion process or the number of feature values ​​that have not yet been excluded by the exclusion process. For example, the feature value selector 11 determines that n increases when the number of valid dimensions (which is the number of feature values ​​remaining after the exclusion process) is equal to or greater than a predetermined threshold, and does not increase when the number of valid dimensions is less than the threshold.Alternatively, feature selector 11 determines that n increments when the number of valid dimensions is greater than a predetermined threshold, and does not increment when the number of valid dimensions is equal to or less than the threshold. In another instance, feature selector 11 determines that n increments when the number of feature values ​​excluded by the exclusion process is equal to or less than a predetermined threshold, and does not increment when the number of feature values ​​is greater than the threshold. Alternatively, feature selector 11 determines that n increments when the number of feature values ​​excluded by the exclusion process is less than a predetermined threshold, and does not increment when the number of feature values ​​is equal to or greater than the threshold.

[0071] In an example, feature selector 11 determines whether to decrement the exclusion quantity n based on the number of feature values ​​excluded by the exclusion process or the number of feature values ​​not yet excluded by the exclusion process. For example, feature selector 11 determines that n decrements when the number of valid dimensions (which is the number of feature values ​​remaining after the exclusion process) is equal to or less than a predetermined threshold, and does not decrement when the number of valid dimensions is greater than the threshold. Alternatively, feature selector 11 determines that n decreases when the number of valid dimensions is less than a predetermined threshold, and does not decrease when the number of valid dimensions is equal to or greater than the threshold. On the other hand, feature selector 11 determines that n decreases when the number of feature values ​​excluded by the exclusion process is equal to or greater than a predetermined threshold, and does not decrease when the number of feature values ​​is less than the threshold. In another instance, feature selector 11 determines that n decreases when the number of feature values ​​excluded by the exclusion process is greater than a predetermined threshold, and does not decrease when the number of feature values ​​is equal to or less than the threshold. Specification 7 / 24 page 11 CN 122374769 A

[0072] When the exclusion quantity n is changed (as in step S16), the process continues to step S17. In step S17, feature selector 11 changes the exclusion quantity n by a predetermined amount. It is changed to either increase or decrease. The predetermined amount is any positive integer. For example, eigenvalue selector 11 increases the exclusion quantity n by 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. As another example, eigenvalue selector 11 increases the number of eigenvalues ​​excluded by a quantity equal to the initial value n. Alternatively, eigenvalue selector 11 decreases the exclusion quantity n by 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. After step S17, the process continues to step S14. In the repeated step S14, eigenvalue selector 11 excludes n eigenvalues ​​from a plurality of eigenvalues ​​based on a selection criterion for each eigenvalue. In the repeated step S15, eigenvalue selector 11 generates and evaluates a provisional prediction model based on the remaining one or more eigenvalues.The feature selector 11 gradually increases or decreases the exclusion quantity n to calculate the evaluation index of the provisional prediction model for each exclusion quantity n.

[0073] When the exclusion quantity n is not changed (not in step S16), the process continues to step S18. In step S18, the feature selector 11 selects one or more input feature values ​​from a plurality of feature values. In an example, the feature selector 11 determines the final exclusion quantity nFINAL based on the final evaluation index obtained for each exclusion quantity. The feature selector 11 may determine the exclusion quantity that obtains the best final evaluation index as the exclusion quantity nFINAL. For example, the feature selector 11 may determine the exclusion quantity that obtains the highest or lowest final evaluation index as the exclusion quantity nFINAL. As an example, when the mean squared error statistic is the evaluation index, the feature selector 11 may determine the exclusion quantity that obtains the lowest final evaluation index as the exclusion quantity nFINAL. When the correlation coefficient statistic is used as the evaluation metric, the eigenvalue selector 11 can determine the number of exclusions that yields the highest final evaluation metric as the exclusion quantity nFINAL. That is, the eigenvalue selector 11 can determine the number of exclusions with the smallest mean squared error statistic or the number of exclusions with the highest correlation coefficient statistic as the final exclusion quantity nFINAL. Alternatively, the eigenvalue selector 11 can determine one or more exclusion quantities that result in a higher final evaluation metric compared to the case where multiple eigenvalues ​​are used, as the exclusion quantity nFINAL. Alternatively, the eigenvalue selector 11 can determine one or more exclusion quantities that result in a decrease in the final evaluation metric less than a predetermined threshold compared to the case where multiple eigenvalues ​​are used, as the exclusion quantity nFINAL. The eigenvalue selector 11 excludes nFINAL eigenvalues ​​from multiple eigenvalues ​​based on the selection metric for each eigenvalue. When using inter-distribution distances such as Wasserstein distance, the eigenvalue selector 11 excludes nFINAL eigenvalues ​​in descending order of inter-distribution distance and selects the remaining one or more eigenvalues ​​as one or more input eigenvalues. In this case, the feature selector 11 selects one or more feature values ​​whose inter-distribution distance is less than a predetermined reference determined based on the final evaluation metric as one or more input feature values. The predetermined reference may be determined by the user or administrator of the information processing system 10, or may be automatically determined by any technology (such as machine learning).

[0074] As described above, the feature selector 11 excludes a portion of the multiple feature values ​​based on the selection metric for each of the multiple feature values. The feature selector 11 selects one or more feature values ​​remaining after the exclusion process as one or more input feature values.In this example, the feature selector 11 calculates the inter-distribution distance for each of the plurality of feature values, which is the distance between a first probability distribution and a second probability distribution. Then, based on the inter-distribution distance for each of the plurality of feature values, it selects one or more input feature values ​​from the plurality of feature values. In this example, the feature selector 11 calculates the inter-distribution distance for each of the plurality of feature values, which is the distance between a first probability distribution and a second probability distribution. Then, it selects the feature value with the smaller inter-distribution distance for each of the plurality of feature values ​​as the input feature value. As described above, in this example, the feature selector 11 repeats the evaluation process, in which a provisional prediction model is generated and an evaluation index for the provisional prediction model is calculated based on the remaining one or more feature values, while changing the number of feature values ​​excluded from the plurality of feature values ​​based on the selection index for each of the plurality of feature values ​​(repeating steps S14 to S17). The feature selector 11 can perform the evaluation process using cross-validation. The specification, page 8 / 24, CN 122374769 A, describes a feature selector 11 that determines the number of exclusions based on multiple evaluation metrics obtained through repeated evaluation processes (step S18). The feature selector 11 excludes a number of n FINAL feature values ​​from the multiple feature values ​​based on the selection metric for each of the multiple feature values ​​to select one or more input feature values ​​(step S18).

[0075] Referring to FIG9, another example of the process for selecting input feature values ​​will be described. FIG9 shows a flowchart illustrating in detail an example of step S10A. Step S10A can also be described as a modification of step S10. The difference between step S10A and step S10 is that the generation and evaluation of the provisional prediction model are repeated, while changing the number of feature values ​​selected m instead of the number of feature values ​​excluded n. The differences will be described in detail below.

[0076] As with step S10, the feature selector 11 performs the processes of steps S11 and S12.

[0077] In step S13A, the feature value selector 11 initializes the selection quantity m of the feature values. The selection quantity m is the number of feature values ​​selected from the multiple feature values ​​based on the selection index of each of the multiple feature values. When the selection quantity m increases, the initial value of the selection quantity m can be 1, 2 or greater. When the selection quantity m decreases, the initial value of the selection quantity m can be the total number of feature values ​​or a smaller number.

[0078] In step S14A, the feature value selector 11 selects m feature values ​​from the multiple feature values ​​based on the selection index of each feature value.When using inter-distribution distances such as Wasserstein distance, the eigenvalue selector 11 selects n eigenvalues ​​in ascending order of inter-distribution distances. That is, the eigenvalue selector 11 selects m eigenvalues ​​where the distance between the first probability distribution and the second probability distribution is relatively small.

[0079] In step S15A, the eigenvalue selector 11 generates and evaluates a provisional prediction model based on one or more selected eigenvalues. That is, the eigenvalue selector 11 performs an evaluation process. “One or more selected eigenvalues” is essentially the same as “one or more remaining eigenvalues” in step S10. Therefore, in step S15A, the cross-validation shown in FIG8 can be performed.

[0080] In step S16A, the eigenvalue selector 11 determines whether to change the selection quantity m. The change in the selection quantity m can be incremented or decremented.

[0081] In this example, the eigenvalue selector 11 determines whether to increment the selection quantity m based on the number of selected eigenvalues ​​or the number of eigenvalues ​​that have not yet been selected. In another instance, feature selector 11 determines whether the selection quantity m is decreased based on the number of selected feature values ​​or the number of unselected feature values.

[0082] When the selection quantity m is changed (as in step S16A), the process continues to step S17A. In step S17A, feature selector 11 changes the selection quantity m by a predetermined amount. This is either increased or decreased. The predetermined amount is any positive integer. For example, feature selector 11 increases the selection quantity m by 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. As another instance, feature selector 11 increases the number of feature values ​​selected by an amount equal to the initial value m. Alternatively, feature selector 11 decreases the selection quantity m by 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. After step S17A, the process continues to step S14A. In the repeated step S14A, the feature selector 11 selects m feature values ​​from a plurality of feature values ​​based on a selection metric for each feature value. In the repeated step S15A, the feature selector 11 generates and evaluates a provisional prediction model based on one or more selected feature values. The feature selector 11 gradually increases or decreases the number of selections m to calculate an evaluation metric for the provisional prediction model for each number of selections m.

[0083] When the number of selections m is not changed (not in step S16A), the process continues to step S18. In step S18, the feature selector 11 selects one or more input feature values ​​from a plurality of feature values. In this example, the feature selector 11 determines the final number of selections mFINAL based on the final evaluation metric obtained for each number of selections.Feature selector 11 selects mFINAL feature values ​​from a plurality of feature values ​​based on a selection metric for each feature value. When using inter-distribution distances such as Wasserstein distance, feature selector 11 selects mFINAL feature values ​​as input feature values ​​in ascending order of inter-distribution distances. That is, feature selector 11 selects one or more feature values ​​whose inter-distribution distance is less than a predetermined reference determined based on a final evaluation metric as one or more input feature values. As mentioned above, the predetermined reference can be determined manually or automatically.

[0084] As mentioned above, feature selector 11 repeats the evaluation process in which the generation of a provisional prediction model and the calculation of an evaluation metric for the provisional prediction model are performed based on the selected one or more feature values, while changing the number of feature values ​​selected from the plurality of feature values ​​based on the selection metric for each of the plurality of feature values ​​(repeating steps S14A to S17A). Feature selector 11 can perform the evaluation process by cross-validation. Feature selector 11 selects one or more input feature values ​​based on multiple evaluation metrics obtained through repeated evaluation processing.

[0085] (Learning Phase)

[0086] The process for generating a prediction model 30 based on the selected one or more input feature values ​​will be described, namely step S20. Learner 12 generates prediction model 30 by performing machine learning (supervised learning) using one or more input feature values ​​of the training data obtained by feature selector 11. In machine learning, feature selector 11 does not use the excluded nFINAL feature values ​​and inputs one or more feature values ​​of the training data as input parameters (e.g., input vectors) into the machine learning model. In an example, feature selector 11 updates the parameter set in the machine learning model by performing backpropagation (backpropagation method) based on the error between the predicted value calculated by the machine learning model and the correct solution (label). Learner 12 obtains prediction model 30 by repeating this process until a given termination condition is met. The termination condition may be processing all data records (first stubs) of the training data. It should be noted that the generated prediction model 30 is assumed to be the optimal computational model, and is not necessarily the "actually optimal computational model".

[0087] (Prediction Stage)

[0088] The prediction performed using the prediction model 30, i.e., step S30, will be described. The predictor 13 obtains one or more input feature values ​​for each of one or more molecules of interest. Each molecule of interest can be a molecule specified by a user of the information processing system 10 through an input operation or a selection operation.Predictor 13 can read input feature values ​​from a predetermined database or receive input feature values ​​from another computer (such as a user terminal) for each molecule of interest. Predictor 13 inputs one or more input feature values ​​of the molecule of interest into prediction model 30. When one or more other feature values ​​are obtained in addition to one or more input feature values ​​of the molecule of interest, predictor 13 does not use one or more other feature values ​​and inputs one or more input feature values ​​as input parameters (e.g., input vectors) into prediction model 30. Predictor model 30 calculates feature values ​​of the molecule of interest based on the input feature values, and predictor 13 obtains feature values ​​from prediction model 30. Therefore, predictor 13 uses prediction model 30 to obtain feature values ​​for each of one or more molecules of interest.

[0089] Predictor 13 can output feature values ​​for each of one or more molecules of interest as a prediction result. When obtaining feature values ​​for each of a plurality of molecules of interest, predictor 13 can select at least one of the plurality of molecules of interest based on the feature values. For example, predictor 13 selects molecules of interest whose feature values ​​satisfy a predetermined reference. Such selection can also be described as screening. Predictor 13 can output information about at least one selected molecule of interest as a prediction result. The information may include at least one of the name, structure, sequence information, and characteristic values ​​of the molecule of interest.

[0090] In step S30, predictor 13 can process at least one of a plurality of second molecules as the molecule of interest. Predictor 13 can acquire one or more feature values ​​for each molecule of interest and output the characteristic value of the molecule of interest obtained by inputting one or more acquired input feature values ​​into the generated prediction model for each molecule of interest. When acquiring the characteristic value for each of the plurality of second molecules, predictor 13 can select at least one of the plurality of second molecules as a candidate molecule based on the characteristic value and output information about the selected at least one candidate molecule.

[0091] Predictor 13 can store the prediction result in a storage device (such as auxiliary memory 103), display the prediction result on an output device 106, or send the prediction result to another computer (such as a user terminal).

[0092] (Method for generating molecular compounds)

[0093] Based on information output during the prediction phase regarding at least one molecule of interest or candidate molecule, a molecular compound having a molecular sequence of at least one molecule of interest or candidate molecule can be generated.

[0094] When the molecular compound is an antibody, the antibody can be produced by, for example, the recombinant method or configuration described in U.S. Patent No. 4,816,567.Examples of methods for generating antibodies include culturing host cells containing nucleic acids encoding the antibody under conditions suitable for expressing the antibody, which is a candidate molecular compound described herein, or collecting the antibody from the host cells (or host cell culture medium). The isolated nucleic acid encoding the antibody may encode an amino acid sequence of the VL containing the antibody and / or an amino acid sequence of the VH containing the antibody (e.g., the light chain and / or heavy chain of the antibody). The host cell containing such nucleic acid contains (1) a vector containing nucleic acids encoding the amino acid sequence of the VL containing the antibody and the amino acid sequence of the VH containing the antibody, or (2) a first vector containing a nucleic acid encoding the amino acid sequence of the VL containing the antibody and a second vector containing a nucleic acid encoding the amino acid sequence of the VH containing the antibody (e.g., the host cell is transformed). In examples, the host cell is a eukaryotic cell (e.g., Chinese hamster ovary (CHO) cell) or a lymphocyte (e.g., Y0, NSO, Sp2 / O cells). Host cells suitable for cloning or expressing vectors encoding antibodies include prokaryotic or eukaryotic cells. For example, bacteria can be used to produce antibodies, especially when glycosylation and Fc effector function are not required. Antibody fragments and peptides can be expressed in bacteria, for example, with reference to US 5648237, US 5789199, and US 5840523. Furthermore, antibody fragments can be expressed in *E. coli*, with reference to “Charlton, Methods in Molecular Biology, Vol. 248 (BKC Lo, ed., Humana Press, Totowa, NJ, 2003), pp. 245-254”. After expression, the antibody can be separated from the bacterial cell paste into a soluble fraction or further purified.

[0095] When the molecular compound is a peptide or cyclic peptide, the compound can be prepared by liquid-phase synthesis, solid-phase synthesis involving Fmoc synthesis, Boc synthesis, etc., or combinations thereof. Solid-phase synthesis is a method in which a compound is bound to a solid and the compound is chemically reacted with a reagent on a solid resin to synthesize the compound of interest. Solid-phase synthesis of peptides is a method in which a desired amino acid or peptide is bound to a solid resin and conjugated with an amino acid or peptide bound to the solid resin, thereby sequentially linking the desired amino acid or peptide to extend the peptide chain, thus synthesizing the peptide. The desired peptide is obtained by separating the peptide bound to the solid resin from the solid resin.

[0096] [Molecule]

[0097] A molecule can be a small molecule (low molecular weight compound), a medium molecular weight compound (medium molecular weight compound), or a large molecule (macro molecular weight compound).In this disclosure, small or low molecular weight compounds are compounds having a molecular weight of less than 500 g / mol. In this disclosure, medium molecular weight compounds are compounds having a molecular weight of 500 g / mol or greater and less than 30,000 g / mol. In this disclosure, high molecular weight compounds are compounds having a molecular weight of 30,000 g / mol or greater.

[0098] Molecules can be biomolecules or non-biomolecules. Molecules can be nucleic acids, peptides, cyclic peptides, proteins, or antigen-binding molecules such as antibodies, or molecules that bind to a target molecule (target-binding molecules). Molecules can be drug candidate molecules. When the molecule is a peptide, cyclic peptide, protein, or antibody, the constituent units of the molecule are amino acids. When the molecule is a nucleic acid, the constituent units of the molecule are nucleosides or nucleotides.

[0099] In this disclosure, the desired properties are those required for a new target of a drug candidate and can be arbitrarily set. Examples of properties include, but are not limited to, the ability to bind to a predetermined in vivo target, pharmacological activity, physical properties, kinetics, and safety. Examples of physical properties include thermal stability, chemical stability, solubility, viscosity, light stability, long-term storage stability, nonspecific adsorption, lipophilicity, and membrane permeability. As an example, when the designed molecule is mRNA (messenger RNA), the property is the ability to translate proteins. A molecule is a molecule that binds to a target molecule, such as an antigen-binding molecule, and the property can be the ability to bind to the target molecule.

[0100] - Antigen-binding molecule

[0101] In this disclosure, antigen-binding molecules containing an antigen-binding domain are used in the broadest sense. Specifically, antigen-binding molecules include various types of molecules as long as they contain an antigen-binding domain. An antigen-binding molecule can be a molecule consisting only of an antigen-binding domain, or a molecule containing an antigen-binding domain and another domain. If the antigen-binding molecule is a molecule formed by an antigen-binding domain and an Fc region bound together, examples of such molecules include complete antibodies and antibody fragments. Antibodies may include monoclonal antibodies (including agonist and antagonist antibodies), human antibodies, humanized antibodies, and chimeric antibodies. The antigen-binding molecules in this disclosure may include scaffold molecules formed by using only a portion of the three-dimensional structure serving as the scaffold (such as existing stable α / β barrel protein structures) as a library for constructing the antigen-binding domain.

[0102] - Binding Capacity Assessment

[0103] The techniques for assessing the ability of a target molecule-binding molecule to bind to a target molecule are not limited. The assessment of binding capacity can be performed by quantitatively evaluating the binding of the target molecule-binding molecule to the target molecule. The target molecule is, for example, a target protein. The target molecule-binding molecule is, for example, an antigen-binding molecule, and the target molecule is, for example, an antigen.For example, when the target molecule is an antigen, it can be assessed by measuring the binding activity of an antigen-binding molecule to the antigen. “Binding activity” is the total strength of non-covalent interactions between one or more binding sites of a molecule (e.g., an antibody) and its binding partner (e.g., an antigen). Here, “binding activity” is not strictly limited to 1:1 interactions between members of a binding pair (e.g., antibody and antigen). For example, when the members of a binding pair reflect a monovalent 1:1 interaction, binding activity means intrinsic binding affinity (sometimes simply referred to as “affinity”). If the members of a binding pair are capable of both monovalent and multivalent binding, then binding activity is the sum of their binding forces. The binding activity of molecule X with its partner Y is typically expressed by the dissociation constant (KD) or “the amount of analyte bound per unit amount of ligand.” For example, an octet value is one indicator of binding capacity and is measured as the amount of analyte bound per unit amount of ligand. Binding activity can be measured using any of the conventional methods known in the art, including those described herein. Those skilled in the art can appropriately determine conditions other than the concentration of compounds specific to the target tissue.

[0104] In one mode, for antibody binding activity, a ligand capture method is used, for example, using BIACORE® T200 or BIACORE® 4000 (GE Healthcare, Uppsala, Sweden), which uses surface plasmon resonance analysis as the measurement principle.

[0105] In another mode, the measurement results are analyzed using BIACORE® evaluation software. The calculation of kinetic parameters is performed by simultaneously fitting the sensor maps of association and dissociation using a 1:1 binding model. Through this process, the association rate (kon or ka), dissociation rate (koff or kd), and equilibrium dissociation constant (KD) can be calculated.

[0106] As a value for antigen binding activity, KD (dissociation rate constant) can be used when the antigen is a soluble molecule, and apparent kd (apparent dissociation rate constant) can be used when the antigen is a membrane-associated molecule. kd (dissociation rate constant) and apparent KD (apparent dissociation rate constant) can be measured by methods known to those skilled in the art. For example, Biacore (GE Healthcare), flow cytometry, etc. can be used.

[0107] Another characterization modality is, for example, using selection techniques for antigen-binding molecules using a displayed library. In one modality, an example is panning using phage display.For example, in affinity assessment, phages presenting antigen-binding molecules that interact with the target antigen (see page 12 / 24, CN 122374769 A) can be concentrated through a process in which a phage library presenting multiple different antigen-binding molecules is prepared, the target antigen is contacted with the prepared phage, and then unbound phages are washed out. Sequences with affinity for the target antigen can be identified by analyzing the nucleic acid sequences encoding the antigen-binding molecules contained in the concentrated phage. An example of this is panning using mammalian cell display. In pharmacological activity assessment using a display system, cells containing genes for antigen-binding molecules with the desired pharmacological activity can be isolated through a process such as flow cytometry, in which a library containing multiple different antigen-binding molecules is expressed in the targeted mammalian cells, and the reporter gene activity is altered based on its effects on the cells. In physical property assessments using a display system, cells possessing genes for antigen-binding molecules capable of stable high-level expression can be isolated using a process such as flow cytometry, in which a library containing multiple different antigen-binding molecules is expressed in targeted mammalian cells and the expression level is examined by staining with an antibody specific to that antigen-binding molecule. Characterization of antigen-binding molecules by panning is not limited to techniques using bacteriophages or mammalian cells, and various techniques can be used as long as they allow the presentation of antigen-binding molecules. Examples include techniques that allow ribosome presentation, techniques that allow mRNA presentation, techniques that allow presentation by viruses other than bacteriophages, and techniques that allow presentation by bacteria (such as E. coli).

[0108] Another characterization modality is, for example, a method for obtaining antibody gene sequences from immune cells derived from an individual or a method for obtaining antibody protein sequences from serum. In affinity assessments involving the extraction of antibody gene sequences from immune cells, sequences with affinity for the target antigen can be identified by a process involving: administering the target antigen protein to an individual to induce immune sensitization, and extracting the antibody gene from an immune cell possessing the gene to obtain an antibody that binds to the target antigen.

[0109] For antigens that induce immune sensitization, techniques using not only proteins but also genes encoding proteins or cells expressing proteins can be used.

[0110] Examples of individuals as subjects include, but are not limited to, humans, mice, rats, hamsters, rabbits, monkeys, chickens, camels, llamas, and alpacas.

[0111] Examples of techniques for analyzing nucleic acid sequence or appearance frequencies include, but are not limited to, techniques for analyzing recombinant organisms in which nucleic acid sequences with different antigen-binding molecules are cloned and analyzed by capillary electrophoresis using the Sanger method and techniques for analysis using next-generation sequencers.

[0112] In nucleic acid sequence analysis, the strength of a property can be determined based on its frequency of occurrence. For example, in nucleic acid sequence analysis after concentration, the property of an antigen-binding molecule encoded by a sequence with a high frequency of occurrence can be assessed as high. On the other hand, the property of an antigen-binding molecule encoded by a sequence with a low frequency of occurrence after concentration can be assessed as lower than the property of an antigen-binding molecule encoded by a sequence with a high frequency of occurrence.

[0113] Techniques for obtaining information about antigen-binding molecules derived from display libraries or individuals are applicable to various types of characterization other than those described above.

[0114] - Pharmacological Activity Assessment

[0115] Techniques for assessing molecular pharmacological activity are not limited thereto. Pharmacological activity can be assessed, for example, by measuring the neutralizing activity, agonist activity, or cytotoxic activity exhibited by the molecule. Examples of assessments of cytotoxic activity as a type of pharmacological activity assessment include the assessment of antibody-dependent cell-mediated cytotoxicity (ADCC) activity, complement-dependent cytotoxicity (CDC) activity, T cell-dependent cytotoxicity (TDCC) activity, and antibody-dependent phagocytic cell (ADCP) activity. CDC activity is cytotoxic activity due to the complement system. ADCC activity is the activity in which an immune cell binds to the Fc region of an antigen-binding molecule containing an antigen-binding domain, which binds to a membrane-associated molecule expressed on the cell membrane of a target cell via an Fcγ receptor expressed on the immune cell, and the immune cell damages the target cell. TDCC activity is the activity in which a T cell damages a target cell by bringing the target cell and the T cell into close proximity using a bispecific antibody containing an antigen-binding domain that binds to a membrane-associated molecule expressed on the cell membrane of the target cell and an antigen-binding domain that targets any of the constituent subunits of the T cell receptor (TCR) complex on the T cell, particularly the CD3ε chain. Whether an antigen-binding molecule of interest possesses ADCC, CDC, TDCC, or ADCP activity can be determined by known methods.

[0116] Neutralizing activity is an activity that targets and inhibits a ligand having biological activity against cells (e.g., viruses and toxins). That is, a substance with neutralizing activity binds to a ligand or a receptor to which the ligand binds to inhibit the binding between the ligand and the receptor. A receptor whose binding to a ligand has been inhibited by neutralizing activity will be prohibited from exhibiting the biological activity mediated by that receptor. Neutralizing activity is not limited to the inhibition of ligand-receptor binding, and activity that inhibits the function of a biologically active protein is also understood as neutralizing activity. Examples of protein functions include enzyme activity.

[0117] - Physical Property Assessment

[0118] As examples, the assessment of stability (such as thermal stability, chemical stability, light stability, stability to mechanical stimulation, and long-term storage stability) can be performed by measuring the decomposition, chemical modification, and association of molecules before and after treatments intended for stability assessment (such as heat treatment, exposure to low pH environments, exposure to light, machine stirring, and long-term storage). Examples of measurement methods used to assess stability include techniques involving chromatography, such as ion exchange chromatography or size exclusion chromatography, mass spectrometry, and electrophoresis. Another measurement method may be used.

[0119] Other examples of assessing physical properties include assessing protein solubility by polyethylene glycol precipitation, assessing viscosity by small-angle X-ray scattering, and assessing nonspecific binding based on assessment of binding to the extracellular matrix (ECM).

[0120] Other examples of assessing physical properties include assessment of protein expression levels, assessment of binding to resins or ligands used for purification, and assessment of surface charge.

[0121] - Kinetic Assessment

[0122] Molecular kinetic assessment can be performed by administering the molecule to an animal (such as a mouse, rat, monkey, or dog) and measuring the amount of the molecule in the blood over time after administration. Alternatively, kinetic assessment can be performed by pharmacokinetic (PK) assessment. As a technique for directly assessing PK, the kinetic behavior of a molecule can be predicted from its amino acid sequence by calculating the molecule's surface charge, isoelectric point, etc., using software.

[0123] - Safety Assessment

[0124] Examples of assessing molecular safety include immunogenicity prediction tools (such as ISPRI network-based immunogenicity screening (EpiVax, Inc.), HLA binding of fragment peptides of antigen-binding molecules, MAP (MHC-related peptide proteomics), and detection of T cell epitopes and assessment of immunogenicity involved in assessing T cell growth. Safety assessments can be performed as long as they can be measured by techniques such as using binding with rheumatoid factor (RF), PBMCs, or whole blood to assess immune responses and platelet aggregation.

[0125] In an example, a drug discovery system including information processing system 10 provides a method for generating novel targets with predetermined properties (such as specific physiological activities, e.g., binding to specific proteins). Examples of drugs include potential activators, such as low-molecular-weight drugs, medium-molecular-weight drugs, biologics, cells, nucleic acid drugs, biopharmaceuticals, and other activators. Targets include molecular structures having desired or defined biological activities. Biological activity may, for example, preferentially bind to a specific protein over other proteins.Molecules that are drug candidates include biomolecules and compounds, including a variety of molecules such as nucleic acids, peptides, cyclic peptides, proteins, antibodies, target molecule binding molecules, high molecular weight compounds, medium molecular weight compounds, and low molecular weight compounds. Specification 14 / 24 pages 18 CN 122374769 A

[0126] A drug discovery system may include a device for selecting molecules that interact with a target of a drug agent and a device for generating lead molecules, etc. A drug discovery system may be, for example, an information processing system that includes the disclosure of WO 2020 / 246617.

[0127] A drug discovery system includes a molecular design device. The molecular design device searches for candidates of molecules having desired properties and outputs information about the identified candidate molecules. The drug discovery system uses the information about the candidate molecules that have been output from the molecular design device to select new targets suitable for drug candidates.

[0128] [Validation Examples]

[0129] (First Validation Example)

[0130] Next, a validation example of an information processing system according to the present disclosure will be described.

[0131] In a first validation example, a dataset representing measurements of the binding affinity of each of a plurality of antibody sequences is used to validate the usefulness of the information processing system according to the present disclosure. The dataset includes sequences of bispecific antibodies that recognize MarvelD3 and CD3 as antigens, along with characteristic values ​​of their binding affinity. The subject of the first validation example is to improve the binding affinity of antibodies to MarvelD3, which is assumed to be an antigen. Binary octet values ​​are used as characteristic values ​​of binding affinity.

[0132] MarvelD3 is a tight junction protein with a 4-transmembrane structure. In the first validation example, MarvelD3 is set as a target candidate for an anticancer drug. The development of bispecific antibodies that bridge cancer antigens and antigens on T cells holds promise for applications in cancer therapy. In the first validation example, a learned prediction model is used when a candidate for anti-MarvelD3 with better properties is identified from a lead antibody.

[0133] In the first validation example, the octet values ​​of the antibody sequences are measured multiple times, and the antibody sequences and their octet values ​​obtained through measurement are used as inputs to the prediction model. Batch measurements are performed to obtain binary octet values, typically 100 or fewer antibody sequences, in a single measurement.

[0134] The antibodies used in the measurements are obtained through the following process. First, a plasmid encoding a pre-designed heavy or light chain is provided, and recombinant antibodies are transiently expressed using Expi293F cells. The antibody is removed from the culture supernatant using protein A and dissolved in a buffer solution. The dissolution buffer solution is mixed under reducing conditions to prepare the MarvelD3 / CD3 bispecific antibody. In the preparation, charge repulsion is applied between identical heavy chains to achieve selective heavy chain heterodimerization.The concentration of the antibody in the buffer solution was determined by absorbance at 280 nm. Subsequently, ion-exchange chromatography was performed on the buffer solution containing the bispecific antibody, confirming the presence of the expected MarvelD3 / CD3 antibody.

[0135] The Octet HTX system was used for octet measurements. Extracellular vesicles with CD81 protein and human MarvelD3 on their surface were captured on a sensor chip using an anti-CD81 antibody. After a 600-second baseline step in D-PBS (-) solution containing 0.1% BSA, the association and dissociation responses were measured for 900 seconds and 1,500 seconds, respectively, in the same buffer solution containing 20 nM antibody. The antibody binding capacity was expressed as a wavelength change between the baseline step and the end of the association phase. Measurements were performed at a temperature of 30°C and a vibration rate of 1,000 times per minute during the baseline step, association, and dissociation phases.

[0136] The specific method is shown below. First, 1,144 antibody sequences were extracted as a training sequence set. Next, training data representing the training sequence set was used to prepare a property prediction model for predicting binding affinity. A virtual sequence set containing 196,608 virtual sequences was input into the property prediction model as a prediction sequence set. This virtual sequence set was prepared by combining candidate sequences for amino acid changes and had no measurement value. The prediction sequence set was screened based on the prediction value of the property prediction model for each antibody sequence contained in the prediction sequence set, and it was verified whether antibody sequences with binding affinity could be selected.

[0137] In the preparation of the property prediction model, the protein language model TAPE was used to convert each antibody sequence into a 768-dimensional vector of each of the VH and VL regions, and the 1,536-dimensional vector data obtained by vector combination was defined as the feature value of each antibody sequence.

[0138] To verify the usefulness of the present invention, input feature values ​​for use in generating a prediction model were selected from 1536 dimensional feature values ​​using two different methods. When constructing a feature prediction model using the selected feature values, the binding affinity levels of the antibody sequences selected by the model were compared. LightGBM (Optical Gradient Boosting Machine) was used in the first method. In the first method, a classification model was prepared to distinguish between the sequence sets used for training and the sequence sets used for prediction, and the importance of each dimension among the 1536 dimensional feature values ​​was calculated based on the feature importance obtained from the model.In the first method, eigenvalues ​​are selected such that the number of eigenvalues ​​with low importance (i.e., eigenvalues ​​that estimate small differences between the training and prediction sequence sets) is increased by 10 x 10 sequentially. In the second method, the inter-distribution distance is defined by the Wasserstein distance of each of the eigenvalues ​​in the dimensions of the training and prediction sequence sets. In the second method, eigenvalues ​​are selected such that the number of eigenvalues ​​with small Wasserstein distances (i.e., eigenvalues ​​whose distributions are relatively consistent between the training and prediction sequence sets) is increased sequentially. A cross-validation method using mean squared error as an indicator is applied to each eigenvalue set selected progressively by each of the first and second methods to determine the eigenvalue set with the highest prediction performance for the training sequence sets. Using the eigenvalue sets, the performance of the property predictions for binding forces using LightGBM is evaluated.

[0139] Figure 10 shows the distribution of measurements of antibody sequences selected as the top 16 antibody sequences when selected from a group of sequences used for prediction in descending order of predicted binding strength for each of the following: wherein no feature value selection is performed and all feature values ​​are used as input feature values ​​(baseline), wherein feature values ​​are selected using feature importance obtained from a class classification model (classifier), and wherein feature values ​​are selected using Wasserstein distance. Figure 10 shows that, as compared to using all feature values ​​and using feature values ​​selected using feature importance, feature values ​​selected using Wasserstein distance can screen for antibody sequences with higher binding strength measurements. That is, Figure 10 shows that the information processing system (feature value selection technique) according to this disclosure is useful.

[0140] (Second Validation Example)

[0141] Next, a second validation example will be described. In the second verification embodiment, the usefulness of the information processing system according to this disclosure using AqSoIDB was demonstrated. AqSoIDB provided a dataset of solubility for 9,982 low-molecular-weight compounds, which was downloaded using Therapeutics Data Common, a platform for providing the ability to download public datasets. A property prediction model for predicting solubility was prepared using 1,000 molecules randomly extracted from all compounds as the first molecular group. Next, the remaining 8,982 molecules were used as the second molecular group, and the second molecular group was selected based on the predicted values ​​of the property prediction model. Subsequently, it was verified whether molecules with high solubility measurements could be selected. For verification, the number of random types was set to 10 when randomly extracting molecules, and 10 verifications were performed. In the verification, feature values ​​were selected according to step S10A.In step S11, training and test data are acquired, including 2,048-dimensional feature values ​​extracted using MolCLR (Molecular Contrastive Learning of Representations via Graph Neural Networks). In steps S12 and S14A, Wasserstein distance and the model's feature importance for classification between the first and second molecular groups are used, respectively, and the two types of selection metrics are compared. In step S13A, the initial value of the number of feature selections, m, is set to 18. In step S15A, cross-validation is used as a method to evaluate the performance of the provisional prediction model, mean squared error is used as a performance evaluation metric, and LightGBM (Optical Gradient Boosting Machine) is used as a supervised learning technique. In step S16A, if the number of feature selections, m, is less than 2,048, the process continues to step S17A; if the number of selections, m, is 2,048, the process continues to step S18. In step S17A, the number of selected features m is incremented by 10. (See page 16 / 24 of specification, CN 122374769 A). In step S18, the set of features with the smallest mean square error of the provisional prediction model is selected as the input features.

[0142] Figure 11 shows the distribution of the average values ​​of the measured values ​​of the compounds selected as the top 100 compounds when selected from a descending order of predicted solubility values ​​for each of the following: where no feature selection is performed and all features are used as input features (baseline); where feature importance is used to select features as a selection criterion (classifier); and where Wasserstein distance is used to select features as a selection criterion (Wasserstein). Figure 11 shows that using Wasserstein distance to select features can screen for compounds with higher solubility measurements compared to using all features and using features selected using feature importance. That is, Figure 11 shows that the information processing system (feature selection technique) according to this disclosure is useful.

[0143] [Variations]

[0144] The technology according to this disclosure has been described in detail above based on various embodiments thereof. However, this disclosure is not limited to the above examples. Various variations can be made to the technology according to this disclosure without departing from the spirit of this disclosure.

[0145] In the above examples, the information processing system 10 includes a learner 12 and a predictor 13 in addition to the feature value selector 11, but the information processing system may not have at least one of the learner and predictor.Predictive models can be ported between computer systems. Therefore, a predictive model generated by an information processing system can be used in another computer system. Alternatively, another computer system can generate a predictive model using machine learning with one or more input feature values ​​selected by the information processing system, where the information processing system uses the predictive model to perform a prediction phase.

[0146] The information processing system can be constructed as a server in a client-server system or implemented in a standalone computer. Alternatively, the information processing system can be implemented in a user terminal that can access a database for storing training and test data via a communication network.

[0147] The process of the method executed by at least one processor is not limited to the examples described above. For example, a step or part of the above process can be omitted, and the steps can be performed in a different order. Any two or more of the above steps can be combined, and a part of the steps can be modified or deleted. Alternatively, other steps can be performed in addition to the steps described above.

[0148] Comparing the magnitudes of two numerical values ​​in this disclosure can use either of two criteria (i.e., "more than" and "more than"), and can use either of two criteria (i.e., "or less than" and "less than").

[0149] In this disclosure, the phrase “at least one processor performs a first process, performs a second process, ... and performs an nth process”, or equivalent wording, indicates a concept including a situation in which the processor performs n processes from the first to the nth process, with changes occurring in between. That is, such wording indicates a concept including both a situation in which all n processes are performed by the same processor and a situation in which the n processes are changed due to any strategy.

[0150] In this disclosure, the term “to” indicates a range including values ​​at both ends. For example, “A to B” means a range of A or greater and B or less.

[0151] In this disclosure, the term “about”, when used in conjunction with a numerical value, means a range between +10% and -10% of that numerical value.

[0152] In this specification, the term “and / or” is used to indicate the matters described before and after “and / or”, or any combination thereof. For example, “A, B and / or C” includes the subject “A”, “B” and “C”, as well as combinations “A and B”, “A and C”, “B and C” and “A and B and C”.

[0153] [Supplement] Specification 17 / 24 pages 21 CN 122374769 A

[0154] It can be understood from the various examples above that this disclosure includes the following aspects.

[0155] (Supplement 1)

[0156] An information processing system includes at least one processor, wherein the at least one processor performs the following processes:

[0157] For each of a plurality of first molecules, acquiring training data representing a plurality of feature values ​​associated with the first molecule;

[0158] For each of a plurality of second molecules, acquiring test data representing a plurality of feature values ​​associated with the second molecule;

[0159] For each of the plurality of feature values, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being a probability distribution in the training data and the second probability distribution being a probability distribution in the test data; and

[0160] Based on the index of each of the plurality of feature values, selecting one or more of the plurality of feature values ​​as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecule property values ​​based on machine learning.

[0161] (Supplement 2)

[0162] According to the information processing system of Supplement 1, the at least one processor performs the following processing:

[0163] For each of the plurality of feature values, calculate the inter-distribution distance as the index, the inter-distribution distance being the distance between the first probability distribution and the second probability distribution; and

[0164] Select one or more input feature values ​​from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values.

[0165] (Supplement 3)

[0166] According to the information processing system of Supplement 2, the at least one processor selects one or more feature values ​​in which the inter-distribution distance is less than a predetermined reference as the one or more input feature values.

[0167] (Supplement 4)

[0168] According to the information processing system of Supplement 2 or 3, the at least one processor calculates the Wasserstein distance as the inter-distribution distance.

[0169] (Supplement 5)

[0170] The information processing system according to any one of Supplements 2 to 4, wherein the at least one processor performs the following processing:

[0171] excluding a portion of the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values, and selecting the remaining one or more feature values ​​as the one or more input feature values.

[0172] (Supplement 6)

[0173] According to the information processing system of Supplement 5, wherein the at least one processor performs the following processes:

[0174] Repeated evaluation process, in which the number of feature values ​​excluded from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values ​​is changed, while generating a provisional prediction model using training data through the machine learning and calculating an evaluation index for the provisional prediction model based on the remaining one or more feature values;

[0175] Determining the number of feature values ​​excluded from the plurality of feature values ​​as the exclusion quantity based on the plurality of evaluation indexes; and Specification 18 / 24 Page 22 CN 122374769 A

[0176] Excluding the exclusion quantity of feature values ​​from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values, and selecting the one or more input feature values.

[0177] (Supplement 7)

[0178] According to the information processing system described in Supplement 6, wherein the plurality of feature values ​​are N feature values, where N is a natural number greater than 2, and

[0179] the at least one processor performs the following processing:

[0180] setting the number of feature values ​​to be excluded from the N feature values ​​to n, where n is a natural number;

[0181] excluding n feature values ​​from the plurality of feature values ​​based on the inter-distribution distance for each of the N feature values;

[0182] performing an evaluation process in which the generation of a first provisional prediction model using the training data through the machine learning and the calculation of an evaluation index for the first provisional prediction model are performed based on (N-n) feature values;

[0183] further excluding n feature values ​​from the (N-n) feature values ​​based on the inter-distribution distance for each of the (N-n) feature values; and

[0184] performing an evaluation process in which (N-2n) feature values ​​are excluded based on the inter-distribution distance for each of the (N-2n) feature values; The feature values ​​are used to perform the generation of a second provisional prediction model using the training data through the machine learning and the calculation of an evaluation index for the second provisional prediction model.

[0185] (Supplement 8)

[0186] According to the information processing system of Supplement 6 or 7, the at least one processor determines the number of times the highest evaluation index is obtained as the exclusion number.

[0187] (Supplement 9)

[0188] According to the information processing system of Supplement 6 or 7, the at least one processor determines the number of times the lowest evaluation index is obtained as the exclusion number.

[0189] (Supplement 10)

[0190] According to the information processing system described in Supplement 5,

[0191] wherein the at least one processor performs the following processes:

[0192] repeating the evaluation process, in which the number of feature values ​​selected from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values ​​is changed, while generating a provisional prediction model using training data through the machine learning and calculating an evaluation metric for the provisional prediction model based on the selected one or more feature values;

[0193] determining the number of feature values ​​selected from the plurality of feature values ​​as the selection quantity based on the plurality of evaluation metrics; and

[0194] selecting the selected number of feature values ​​as the one or more input feature values ​​based on the inter-distribution distance for each of the plurality of feature values.

[0195] (Supplement 11)

[0196] According to the information processing system described in Supplement 10,

[0197] wherein the plurality of feature values ​​are N feature values, where N is a natural number greater than 2,

[0198] the at least one processor performs the following processing:

[0199] setting the number of feature values ​​selected from the N feature values ​​to m, where m is a natural number; Specification 19 / 24 Page 23 CN 122374769 A

[0200] selecting m feature values ​​from the plurality of feature values ​​based on the inter-distribution distance for each of the N feature values;

[0201] performing an evaluation process, in which the generation of a first provisional prediction model using the training data through the machine learning and the calculation of an evaluation index for the first provisional prediction model are performed based on the m feature values;

[0202] further adding the m feature values ​​to the m feature values ​​based on the inter-distribution distance for each of the m feature values; and

[0203] An evaluation process is performed, in which a second provisional prediction model is generated using the training data through the machine learning based on 2m feature values, and an evaluation index for the second provisional prediction model is calculated.

[0204] (Supplement 12)

[0205] According to the information processing system of Supplement 11, the at least one processor determines the number of times the highest evaluation index is obtained as the selection number.

[0206] (Supplement 13)

[0207] According to the information processing system of Supplement 11, the at least one processor determines the number of times the lowest evaluation index is obtained as the selection number.

[0208] (Supplement 14)

[0209] According to the information processing system of any one of Supplements 6 to 13, the at least one processor performs the evaluation process by cross-validation.

[0210] (Supplement 15)

[0211] The information processing system according to any one of Supplements 1 to 14, wherein the at least one processor generates the predictive model by performing the machine learning using the one or more input feature values ​​of the training data.

[0212] (Supplement 16)

[0213] The information processing system according to Supplement 15, wherein the at least one processor performs the following processes:

[0214] acquiring one or more input feature values ​​for a molecule of interest, the characteristic values ​​of which are unknown; and

[0215] outputting the characteristic values ​​of the molecule of interest, which are obtained by inputting the acquired one or more input feature values ​​into the generated predictive model.

[0216] (Supplement 17)

[0217] According to the information processing system of Supplement 15, the at least one processor performs the following processes:

[0218] acquiring one or more input feature values ​​for each of a plurality of molecules of interest, the feature value of each of the plurality of molecules of interest being unknown;

[0219] inputting the one or more input feature values ​​acquired for each of the plurality of molecules of interest into a generated prediction model, and acquiring the feature value of the molecule of interest from the prediction model;

[0220] selecting at least one molecule of interest from the plurality of molecules of interest based on the feature value of each of the plurality of molecules of interest; and

[0221] outputting information about the at least one selected molecule of interest.

[0222] (Supplement 18)

[0223] According to the information processing system of any one of Supplements 1 to 17, each of the first molecule and the second molecule is selected from nucleic acids, peptides, cyclic peptides, proteins, antibodies and low molecular weight compounds.

[0224] (Supplement 19)

[0225] According to the information processing system of Supplement 15, the at least one processor performs the following processes:

[0226] for each of at least one of the plurality of second molecules, acquiring the one or more input feature values; and

[0227] for each of the at least one second molecule, outputting the feature value of the second molecule obtained by inputting the acquired one or more input feature values ​​into the generated prediction model.

[0228] (Supplement 20)

[0229] According to the information processing system of Supplement 15, wherein the at least one processor performs the following processes:

[0230] For each of the plurality of second molecules, acquiring one or more input feature values;

[0231] Inputting the acquired one or more input feature values ​​into a prediction model generated for each of the plurality of second molecules, and acquiring the feature value of the molecule of interest from the prediction model;

[0232] Based on the feature value of each of the plurality of second molecules, selecting at least one second molecule as a candidate molecule from the plurality of second molecules; and

[0233] Outputting information about the at least one selected candidate molecule.

[0234] (Supplement 21)

[0235] According to the information processing system of any one of Supplements 1 to 20, wherein the feature value is selected from at least one of affinity, pharmacological activity, physical properties, kinetics, and safety.

[0236] (Supplement 22)

[0237] According to any one of Supplements 1 to 20,

[0238] wherein the first molecule and the second molecule are antigen-binding molecules, and

[0239] the characteristic value is the value of the ability of the antigen-binding molecule to bind to the antigen.

[0240] (Supplement 23)

[0241] An information processing method, performed by an information processing system including at least one processor, the method comprising the following steps:

[0242] For each of a plurality of first molecules, acquiring training data representing a plurality of feature values ​​associated with the first molecule;

[0243] For each of a plurality of second molecules, acquiring test data representing the plurality of feature values ​​associated with the second molecule;

[0244] For each of the plurality of feature values, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being a probability distribution in the training data and the second probability distribution being a probability distribution in the test data; and

[0245] Based on the index of each of the plurality of feature values, selecting one or more of the plurality of feature values ​​as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecule property values ​​based on machine learning.

[0246] (Supplement 24)

[0247] An information processing program that enables a computer to perform the following steps:

[0248] For each of a plurality of first molecules, acquiring training data representing a plurality of feature values ​​associated with the first molecule; Specification 21 / 24 pages 25 CN 122374769 A

[0249] For each of a plurality of second molecules, acquiring test data representing the plurality of feature values ​​associated with the second molecule;

[0250] For each of the plurality of feature values, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being a probability distribution in the training data and the second probability distribution being a probability distribution in the test data; and

[0251] Based on the index of each of the plurality of feature values, selecting one or more of the plurality of feature values ​​as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecule property values ​​based on machine learning.

[0252] (Supplement 25)

[0253] A method for generating a molecular compound, the method comprising a generation step of generating a molecular compound having a molecular sequence of the at least one molecule of interest based on information output from an information processing system according to Supplement 17.

[0254] (Supplement 26)

[0255] A method for generating a molecular compound, the method comprising a generation step of generating a molecular compound having a molecular sequence of the at least one candidate molecule based on information output from an information processing system according to Supplement 20.

[0256] (Supplement 27)

[0257] A non-transitory and computer-readable recording medium configured to store an information processing program that enables a computer to perform the following steps:

[0258] For each of a plurality of first molecules, acquiring training data representing a plurality of feature values ​​associated with the first molecule;

[0259] For each of a plurality of second molecules, acquiring test data representing the plurality of feature values ​​associated with the second molecule;

[0260] For each of the plurality of feature values, calculating an index based on a first probability distribution and a second probability distribution, the first probability distribution being a probability distribution in the training data and the second probability distribution being a probability distribution in the test data; and

[0261] Based on the index of each of the plurality of feature values, selecting one or more of the plurality of feature values ​​as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecule property values ​​based on machine learning.

[0262] As described in Supplements 1, 23, 24 and 27, the feature values ​​to be used as input features for the machine learning-based prediction model are selected based on the trend of each feature value in both the training data and the test data.By selecting feature values ​​as described above, the accuracy of the machine learning model (prediction model) used to predict the properties of molecules can be improved. Improving the accuracy of the machine learning model can lead to accurate prediction of the properties of molecules.

[0263] According to Supplement 2, the feature values ​​are selected based on the inter-distribution distance between the first probability distribution and the second probability distribution. By introducing the inter-distribution distance, which represents the quantitative difference in the trend between the two probability distributions, the input feature values ​​can be selected based on objective criteria. Therefore, a further improvement in the accuracy of the machine learning model can be expected.

[0264] According to Supplement 3, feature values ​​with relatively small inter-distribution distances are selected as input feature values. The input feature values ​​selected as described above are relatively consistent in trend between the training data and the test data. Therefore, the accuracy of the machine learning model (prediction model) can be improved by machine learning based on the input feature values. Specification 22 / 24 pages 26 CN 122374769 A

[0265] According to Supplement 4, the inter-distribution distance can be appropriately calculated by introducing the Wasserstein distance as the inter-distribution distance. Therefore, the input feature values ​​can be selected more appropriately.

[0266] According to Supplement 5, input feature values ​​can be selected by focusing attention on feature values ​​that are evaluated as unsuitable as input feature values.

[0267] According to Supplement 6, the generation and evaluation of the provisional prediction model are repeated, while varying the number of excluded feature values, and the number of exclusions is dynamically determined based on multiple evaluation metrics obtained through repetition. The determined number of excluded feature values ​​is excluded to select input feature values. In an example, if too many feature values ​​with relatively large inter-distribution distances are excluded, even feature values ​​that actually contribute to predicting feature values ​​can be excluded. By dynamically determining the number of exclusions as described above, one or more input feature values ​​that improve the accuracy of the machine learning model can be appropriately selected.

[0268] According to Supplement 7, the rate of increase in the number of exclusions is constant during the repetition of the evaluation process, and therefore the evaluation process can be repeated efficiently using convenient techniques.

[0269] According to Supplement 8, the number of exclusions that yields the highest evaluation metric is determined as the final number of exclusions. According to Supplement 9, the number of exclusions that yields the lowest evaluation metric is determined as the final number of exclusions. Thus, one or more input feature values ​​can be selected, with the expectation that these one or more input feature values ​​will give the machine learning model the highest accuracy.

[0270] According to Supplement 10, the generation and evaluation of the provisional prediction model are repeated, while changing the number of selected feature values ​​and dynamically determining the number of selections based on multiple evaluation metrics obtained through the repetition. The determined number of selections is used as input for feature values. In the example, if too many feature values ​​with relatively large inter-distribution distances are selected, even the number of feature values ​​that actually contribute to predicting the feature values ​​can be selected.By dynamically determining the number of selections as described above, one or more input feature values ​​can be appropriately selected to improve the accuracy of the machine learning model.

[0271] According to Supplement 11, the increase in the number of selections is constant during the repetition of the evaluation process, and therefore the evaluation process can be repeated efficiently using convenient techniques.

[0272] According to Supplement 12, the number of selections for the feature value that yields the highest evaluation metric is determined as the final number of selections. According to Supplement 13, the number of selections for the feature value that yields the lowest evaluation metric is determined as the final number of selections. Therefore, one or more input feature values ​​can be selected, with the expectation that the one or more input feature values ​​will give the machine learning model the highest accuracy.

[0273] According to Supplement 14, cross-validation is introduced in the evaluation process for obtaining the evaluation metric, and therefore, even when the amount of training data is limited, each provisional prediction model can be accurately evaluated by effectively utilizing the limited data.

[0274] According to Supplement 15, machine learning for generating a prediction model is performed using one or more input feature values ​​selected based on the trend of each of the feature values ​​in both the training and test data. Therefore, a prediction model with good accuracy can be obtained.

[0275] According to Supplement 16, the properties of a molecule of interest can be predicted with good accuracy by using a prediction model generated by machine learning using one or more selected input feature values.

[0276] According to Supplement 17, the properties of each molecule of interest can be predicted with good accuracy by using a prediction model generated by machine learning using one or more selected input feature values. Therefore, at least one molecule of interest can be appropriately selected from a plurality of molecules of interest.

[0277] According to Supplement 18, the accuracy of the machine learning model (prediction model) used to predict the properties of nucleic acids, peptides, cyclic peptides, proteins, antibodies, and low molecular weight compounds can be improved. Specification 23 / 24 pages 27 CN 122374769 A

[0278] According to Supplement 19, the properties of a second molecule can be predicted with good accuracy by using a prediction model generated by machine learning using one or more selected input feature values.

[0279] According to Supplement 20, the properties of each second molecule can be predicted with good accuracy by using a prediction model generated by machine learning using one or more selected input feature values. Therefore, at least one candidate molecule can be appropriately selected from a plurality of candidate molecules.

[0280] According to Supplement 21, the accuracy of machine learning models (predictive models) used to predict the affinity, pharmacological activity, physical properties, kinetics, or safety of molecules can be improved.

[0281] According to Supplement 22, the accuracy of machine learning models (predictive models) used to predict the ability of antigen-binding molecules to bind to antigens can be improved.

[0282] According to Supplement 25, a molecular compound that can be expected to have desired properties can be generated based on information about at least one molecule of interest appropriately selected using a prediction model.

[0283] According to Supplement 26, a molecular compound that can be expected to have desired properties can be generated based on information about at least one candidate molecule appropriately selected using a prediction model.

[0284] [List of Reference Symbols]

[0285] 10…Information processing system, 11…Eigenvalue selector, 12…Learner, 13…Predictor, 20…Database, 30…Prediction model, 110…Information processing program, 210…First probability distribution, 220…Second probability distribution. Instruction manual, page 24 / 24; Figure 1 (CN 122374769 A); Figure 2 (CN 122374769 A); Figure 3 (CN 122374769 A); Figure 4 (CN 122374769 A); Figure 5 (CN 122374769 A); Figure 6 (CN 122374769 A); Figure 7 (CN 122374769 A); Figure 8 (CN 122374769 A); Figure 9 (CN 122374769 A); Figure 10 (CN 122374769 A). Figure 11, page 38, illustration 10 / 11, CN 122374769 A; Figure 11, page 39, illustration 11 / 11, CN 122374769 A.

Claims

1. An information processing system comprising at least one processor, The at least one processor performs the following processing: For each of the plurality of first molecules, training data representing multiple feature values ​​associated with the first molecule is obtained; For each of the plurality of second molecules, test data representing multiple feature values ​​associated with the second molecule are obtained; For each of the plurality of feature values, an index is calculated based on a first probability distribution and a second probability distribution, wherein the first probability distribution is the probability distribution in the training data and the second probability distribution is the probability distribution in the test data; as well as Based on the index of each of the plurality of feature values, one or more of the plurality of feature values ​​are selected as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecular property values ​​based on machine learning.

2. The information processing system according to claim 1, The at least one processor performs the following processing: For each of the plurality of feature values, the inter-distribution distance is calculated as the index, wherein the inter-distribution distance is the distance between the first probability distribution and the second probability distribution; and One or more input feature values ​​are selected from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values.

3. The information processing system according to claim 2, wherein the at least one processor selects one or more feature values ​​in which the distance between the distributions is less than a predetermined reference as the one or more input feature values.

4. The information processing system according to claim 2 or 3, wherein the at least one processor calculates the Wasserstein distance as the inter-distribution distance.

5. The information processing system according to any one of claims 2 to 4, The at least one processor performs the following processing: A subset of the plurality of feature values ​​is excluded based on the inter-distribution distance for each of the plurality of feature values, and the remaining one or more feature values ​​are selected as the one or more input feature values.

6. The information processing system according to claim 5, The at least one processor performs the following processing: The evaluation process involves altering the number of feature values ​​excluded from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values, while simultaneously generating a provisional prediction model using the training data through the machine learning process and calculating an evaluation metric for the provisional prediction model based on the remaining one or more feature values. Based on multiple evaluation indicators, the number of feature values ​​to be excluded from the multiple feature values ​​is determined as the exclusion quantity; as well as The number of feature values ​​to be excluded is excluded from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values, and one or more input feature values ​​are selected.

7. The information processing system according to claim 6, The plurality of feature values ​​are N feature values, where N is a natural number greater than 2, and The at least one processor performs the following processing: The number of feature values ​​to be excluded from the N feature values ​​is set to n, where n is a natural number; Based on the inter-distribution distance for each of the N feature values, n feature values ​​are excluded from the plurality of feature values; An evaluation process is performed, in which a first provisional prediction model is generated using the training data through the machine learning based on (Nn) feature values, and an evaluation metric for the first provisional prediction model is calculated. Further, based on the inter-distribution distance for each of the (Nn) eigenvalues, n eigenvalues ​​are excluded from the (Nn) eigenvalues; as well as An evaluation process is performed, in which a second provisional prediction model is generated using the training data through the machine learning based on (N-2n) feature values, and an evaluation metric for the second provisional prediction model is calculated.

8. The information processing system according to claim 6 or 7, wherein the at least one processor determines the number of times the highest evaluation index is obtained as the exclusion quantity, or determines the number of times the lowest evaluation index is obtained as the exclusion quantity.

9. The information processing system according to any one of claims 2 to 4, The at least one processor performs the following processing: The evaluation process involves altering the number of feature values ​​selected from the plurality of feature values ​​based on the inter-distribution distance for each of the plurality of feature values, while simultaneously generating a provisional prediction model using the training data through the machine learning process based on one or more selected feature values, and calculating an evaluation metric for the provisional prediction model. Based on multiple evaluation indicators, the number of feature values ​​selected from the multiple feature values ​​is determined as the selection quantity; as well as Based on the inter-distribution distance for each of the plurality of feature values, the selected number of feature values ​​are chosen as the one or more input feature values.

10. The information processing system according to any one of claims 6 to 9, wherein the at least one processor performs the evaluation processing via cross-validation.

11. The information processing system according to any one of claims 1 to 10, wherein the at least one processor generates the predictive model by performing the machine learning using one or more input feature values ​​of the training data.

12. The information processing system according to claim 11, The at least one processor performs the following processing: Obtain one or more input feature values ​​for a molecule of interest, wherein the characteristic values ​​of the molecule of interest are unknown; and The characteristic value of the molecule of interest is output, which is obtained by inputting one or more acquired input feature values ​​into the generated prediction model.

13. The information processing system according to claim 11, The at least one processor performs the following processing: Obtain one or more input feature values ​​for each of a plurality of molecules of interest, wherein the feature values ​​of each of the plurality of molecules of interest are unknown; One or more input feature values ​​obtained for each of the plurality of molecules of interest are input into the generated prediction model, and the characteristic values ​​of the molecules of interest are obtained from the prediction model; At least one molecule of interest is selected from the plurality of molecules of interest based on the characteristic value of each of the plurality of molecules of interest; as well as Output information about at least one selected molecule of interest.

14. The information processing system according to any one of claims 1 to 13, wherein each of the first molecule and the second molecule is selected from nucleic acids, peptides, cyclic peptides, proteins, antibodies and low molecular weight compounds.

15. An information processing method, executed by an information processing system including at least one processor, the information processing method comprising the following steps: For each of the plurality of first molecules, training data representing multiple feature values ​​associated with the first molecule is obtained; For each of the plurality of second molecules, test data representing multiple feature values ​​associated with the second molecule are obtained; For each of the plurality of feature values, an index is calculated based on a first probability distribution and a second probability distribution, wherein the first probability distribution is the probability distribution in the training data and the second probability distribution is the probability distribution in the test data; as well as Based on the index of each of the plurality of feature values, one or more of the plurality of feature values ​​are selected as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecular property values ​​based on machine learning.

16. An information processing program that enables a computer to perform the following steps: For each of the plurality of first molecules, training data representing multiple feature values ​​associated with the first molecule is obtained; For each of the plurality of second molecules, test data representing multiple feature values ​​associated with the second molecule are obtained; For each of the plurality of feature values, an index is calculated based on a first probability distribution and a second probability distribution, wherein the first probability distribution is the probability distribution in the training data and the second probability distribution is the probability distribution in the test data; as well as Based on the index of each of the plurality of feature values, one or more of the plurality of feature values ​​are selected as one or more input feature values ​​to be used as input parameters for a predictive model for predicting molecular property values ​​based on machine learning.

17. A method for manufacturing a molecular compound, the method comprising: The step of generating a molecular compound having the molecular sequence of the at least one molecule of interest based on the information output from the information processing system of claim 13.

Images