Chemical structure proposing method, program and chemical structure proposing device

Abstract

To provide a chemical structure proposing method, a program, and a chemical structure proposing device for efficiently acquiring a candidate structure with polymerizability.SOLUTION: A chemical structure proposing device 10 includes a reactive-end bonding processing part 130 and a generation part 150. The reactive-end bonding processing part 130 executes reactive-end bonding processing. The reactive-end bonding processing is a process of determining a reactive end structure selected from a first group as a reactive end structure to be bonded to a main structure selected from a second group. The first group consists of a plurality of reactive end structures. The second group consists of a plurality of main structures. The generation part 150 generates information representing a bonding structure. The bonding structure is a structure acquired by bonding the main structure selected from the second group and the reactive end structure selected from the first group. The main structure has a plurality of bonding points. The reactive end structure has only one bonding point and contains a polymerizable functional group.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a chemical structure proposing method, a program, and a chemical structure proposing device. [Background technology]

[0002] Attempts to develop materials are being made using computer programs that suggest chemical structures.

[0003] Patent Document 1 describes a method of randomly changing an initial structure and repeatedly estimating the physical properties of the changed structure to obtain a structure having target physical property values.

[0004] Patent Document 2 describes a method for outputting a candidate list by repeating the process of selecting materials based on characteristic values ​​predicted by a prediction model and generating mutant types of the materials.

[0005] Patent document 3 describes calculating a physical property evaluation value and a usability evaluation value of a material based on a chemical formula, and using the physical property evaluation value and the usability evaluation value to select materials to be presented as candidate materials. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] International Publication No. 2020 / 054841 [Patent Document 2] Special Publication No. 2022-516697 [Patent Document 3] JP 2021-174473 A Summary of the Invention [Problem to be solved by the invention]

[0007] In the techniques of Patent Documents 1 to 3, for example, chemical structures are generated by a method of mutating a part of a base structure, etc. In resin development, etc., it is necessary to obtain a candidate structure having polymerizability, but the techniques of Patent Documents 1 to 3 were not able to efficiently obtain a candidate structure having polymerizability.

[0008] The present invention provides a chemical structure proposing method, a program, and a chemical structure proposing device that can efficiently obtain candidate structures having polymerizability. [Means for solving the problem]

[0009] According to one embodiment of the present invention, there are provided the following chemical structure proposing method, program, and chemical structure proposing device.

[0010] 1. One or more computers: performing a reactive end binding process for determining the reactive end structure selected from a first group consisting of a plurality of reactive end structures as the reactive end structure to be bound to the main structure selected from a second group consisting of a plurality of main structures; generating information indicating a bond structure obtained by bonding the main structure selected from the second group and the reactive terminal structure selected from the first group; The main structure has a plurality of attachment points; The reactive end structure has only one attachment point and contains a polymerizable functional group. Chemical structure proposal method. 2. In the chemical structure proposal method according to 1, The one or more computers determine, using a predetermined probability SP1, whether the reactive end structure or the non-reactive end structure is to be bonded to one of a plurality of bonding points of the main structure included in the bond structure; The non-reactive end structure has only one attachment point and does not contain a polymerizable functional group. Chemical structure proposal method. 3. In the chemical structure proposal method according to 2., The one or more computers accept input of information indicating the probability SP1. Chemical structure proposal method. 4. In the chemical structure proposal method according to 2. or 3., The one or more computers select the non-reactive end structure to be bound to the main structure included in the binding structure from a third group consisting of a plurality of the non-reactive end structures. Chemical structure proposal method. 5. In the chemical structure proposal method according to any one of 1. to 4., The one or more computers generate information representing the bond structure including a predetermined number of FN or more of the reactive terminal structures. Chemical structure proposal method. 6. In the chemical structure proposal method according to 5., The one or more computers accept input of information indicative of the number FN. Chemical structure proposal method. 7. In the chemical structure proposal method according to any one of 1. to 6., The one or more computers further perform a main structure combining process for determining, in the combined structure, to combine a first main structure selected from the second group with a second main structure selected from the second group. Chemical structure proposal method. 8. In the chemical structure proposal method according to 7., the one or more computers determine, using a predetermined probability SP2, whether the second main structure or the non-reactive terminal structure is to be bonded to one of a plurality of bonding points of the first main structure in the bonded structure; The non-reactive end structure has only one attachment point and does not contain a polymerizable functional group. Chemical structure proposal method. 9. In the chemical structure proposal method according to 8., The one or more computers accept input of information indicating the probability SP2. Chemical structure proposal method. 10. In the chemical structure proposal method according to any one of 7. to 9., The one or more computers repeat the main structure binding process multiple times. Chemical structure proposal method. 11. In the chemical structure proposal method according to any one of 7. to 10., The one or more computers determine whether or not to perform the main structure joining process using a predetermined probability OP. Chemical structure proposal method. 12. The method for proposing a chemical structure according to 11. The one or more computers receive input of information indicating the probability OP. Chemical structure proposal method. 13. In the chemical structure proposal method according to any one of 1. to 12., The polymerizable functional group is any one of an epoxy group, a hydroxy group, an amino group, an isocyanate group, a mercapto group, a vinyl group, a (meth)acrylic group, a carboxyl group, a formyl group, a norbornene ring, a maleimide ring, and an oxazine ring. Chemical structure proposal method. 14. In the chemical structure proposal method according to any one of 1. to 13., The one or more computers generate a plurality of pieces of information indicating the bond structure. Chemical structure proposal method. 15. In the chemical structure proposal method according to any one of 1. to 14., The one or more computers further estimate properties of the binding structure using a machine learning model. Chemical structure proposal method. 16. The method for proposing a chemical structure according to 15., The characteristic is the specific gravity of a cured resin obtained by using a substance having the bonding structure. Chemical structure proposal method. 17. A program for causing a computer to function as a chemical structure proposal device, comprising: The computer, a reactive end linking processing means for performing a reactive end linking process for determining the reactive end structure selected from a first group consisting of a plurality of reactive end structures as the reactive end structure to be linked to the main structure selected from a second group consisting of a plurality of main structures; and a generating means for generating information indicating a bond structure obtained by bonding the main structure selected from the second group and the reactive terminal structure selected from the first group; Functioning as a The main structure has a plurality of attachment points; The reactive end structure has only one attachment point and contains a polymerizable functional group. program. 18. A reactive end binding processing unit that performs a reactive end binding process to determine the reactive end structure selected from a first group consisting of a plurality of reactive end structures as the reactive end structure to be bound to the main structure selected from a second group consisting of a plurality of main structures; a generating unit that generates information indicating a bond structure obtained by bonding the main structure selected from the second group and the reactive terminal structure selected from the first group; Equipped with The main structure has a plurality of attachment points; The reactive end structure has only one attachment point and contains a polymerizable functional group. Chemical structure proposal device. Effect of the Invention

[0011] According to the present invention, it is possible to provide a chemical structure proposing method, a program, and a chemical structure proposing device that can efficiently obtain candidate structures having polymerizability. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram showing an overview of a chemical structure proposal device according to a first embodiment. [Diagram 2] FIG. 1 is a diagram showing an overview of a chemical structure proposing method according to a first embodiment. [Diagram 3]FIG. 1 is a diagram illustrating a computer for realizing a chemical structure proposal device. [Figure 4] FIG. 1 is a diagram illustrating an example of the functional configuration of a chemical structure proposal device according to a first embodiment. [Diagram 5] 1(a) to 1(e) are diagrams illustrating reactive end structures belonging to the first group. [Figure 6] 11(a) to 11(e) are diagrams illustrating main structures belonging to the second group. [Figure 7] 1(a) to 1(e) are diagrams illustrating examples of non-reactive terminal structures belonging to the third group. [Figure 8] FIG. 2 is a diagram for explaining the concept of a coupling structure according to the first embodiment. [Figure 9] 1 is a flowchart illustrating the flow of processing performed by the chemical structure proposal device according to the first embodiment. [Figure 10] 1 is a flowchart illustrating the flow of a reactive terminal binding process. [Figure 11] FIG. 11 is a diagram illustrating a functional configuration of a chemical structure proposal device according to a second embodiment. [Figure 12] 11A and 11B are diagrams for explaining the concept of a coupling structure according to a second embodiment. [Figure 13] 10 is a flowchart illustrating the flow of processing performed by a chemical structure proposal device according to a second embodiment. [Figure 14] 13 is a flowchart illustrating a flow of a main structure combining process; [Figure 15] 13 is a flowchart illustrating the flow of processing performed by a chemical structure proposal device according to a third embodiment. [Figure 16] FIG. 13 is a diagram illustrating an example of the functional configuration of a chemical structure proposal device according to a fourth embodiment. [Figure 17] 13 is a flowchart illustrating the flow of processing performed by a chemical structure proposal device according to a fourth embodiment. [Figure 18] 1(a) to 1(c) are diagrams showing bond structure 1 to bond structure 3, respectively. [Figure 19]1(a) to 1(c) are diagrams showing the structures of substances B to D, respectively. [Figure 20] FIG. 1 shows the results of identification of substance A. [Figure 21] FIG. 1 shows the results of identification of substance B. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same components are given the same reference numerals and the description will be omitted as appropriate.

[0014] (First embodiment) FIG. 1 is a diagram showing an overview of a chemical structure proposal device 10 according to the first embodiment. The chemical structure proposal device 10 according to the present embodiment includes a reactive terminal bond processing unit 130 and a generation unit 150. The reactive terminal bond processing unit 130 performs a reactive terminal bond process. The reactive terminal bond process is a process for determining a reactive terminal structure selected from a first group as a reactive terminal structure to be bonded to a main structure selected from a second group. The first group is composed of a plurality of reactive terminal structures. The second group is composed of a plurality of main structures. The generation unit 150 generates information indicating a bond structure. The bond structure is a structure obtained by bonding a main structure selected from the second group and a reactive terminal structure selected from the first group. The main structure has a plurality of bond points. The reactive terminal structure has only one bond point and includes a polymerizable functional group.

[0015] FIG. 2 is a diagram showing an overview of the chemical structure proposal method according to the present embodiment. The chemical structure proposal method according to the present embodiment is executed by one or more computers. In the chemical structure proposal method according to the present embodiment, one or more computers perform a reactive end binding process (step S10). The reactive end binding process is a process for determining a reactive end structure selected from the first group as a reactive end structure to be bound to a main structure selected from the second group. The first group is composed of a plurality of reactive end structures. The second group is composed of a plurality of main structures. In the chemical structure proposal method according to the present embodiment, one or more computers generate information indicating a binding structure (step S11). The binding structure is a structure obtained by binding a main structure selected from the second group and a reactive end structure selected from the first group. The main structure has a plurality of binding points. The reactive end structure has only one binding point and includes a polymerizable functional group.

[0016] The chemical structure proposing method according to the present embodiment can be executed by the chemical structure proposing device 10 according to the present embodiment.

[0017] The hardware configuration of the chemical structure proposal device 10 will be described below. Each functional component of the chemical structure proposal device 10 (reactive terminal bond processing unit 130 and generation unit 150) may be realized by hardware that realizes each functional component (e.g., hardwired electronic circuit, etc.), or may be realized by a combination of hardware and software (e.g., a combination of an electronic circuit and a program that controls it, etc.). Below, a further description will be given of the case where each functional component of the chemical structure proposal device 10 is realized by a combination of hardware and software.

[0018] FIG. 3 is a diagram illustrating a computer 1000 for realizing the chemical structure proposal device 10. The computer 1000 is any computer. For example, the computer 1000 is a system on chip (SoC), a personal computer (PC), a server machine, a tablet terminal, or a smartphone. The computer 1000 may be a dedicated computer designed for realizing the chemical structure proposal device 10, or may be a general-purpose computer. The chemical structure proposal device 10 may be realized by one computer 1000 or by a combination of multiple computers 1000.

[0019] The computer 1000 includes a bus 1020, a processor 1040, a memory 1060, a storage device 1080, an input / output interface 1100, and a network interface 1120. The bus 1020 is a data transmission path for the processor 1040, the memory 1060, the storage device 1080, the input / output interface 1100, and the network interface 1120 to transmit and receive data to and from each other. However, the method of connecting the processor 1040 and the like to each other is not limited to bus connection. The processor 1040 is various processors such as a central processing unit (CPU), a graphics processing unit (GPU), or a field-programmable gate array (FPGA). The memory 1060 is a main storage device realized using a random access memory (RAM) or the like. The storage device 1080 is an auxiliary storage device realized using a hard disk, a solid state drive (SSD), a memory card, or a read only memory (ROM) or the like.

[0020] The input / output interface 1100 is an interface for connecting the computer 1000 to an input / output device. For example, an input device such as a keyboard, a mouse, or a touch panel, and an output device such as a display or a speaker are connected to the input / output interface 1100. The input / output interface 1100 may be connected to the input device or the output device by wireless connection or by wired connection.

[0021] The network interface 1120 is an interface for connecting the computer 1000 to a network. This communication network is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network). The method for connecting the network interface 1120 to the network may be a wireless connection or a wired connection.

[0022] The storage device 1080 stores program modules that realize each functional component of the chemical structure proposing device 10. The processor 1040 reads each of these program modules into the memory 1060 and executes them to realize the function corresponding to each program module.

[0023] The chemical structure proposal device 10 according to the present embodiment is a device that proposes a chemical structure by generating information indicating a bond structure. The chemical structure proposed by the chemical structure proposal device 10 may be an existing structure or a new structure. The information indicating the bond structure is hereinafter also referred to as "structure information."

[0024] Since the reactive terminal structure includes a polymerizable functional group, the bond structure shown in the structural information has one or more polymerizable functional groups. The polymerizable functional group may be a polymerizable portion, and its structure is not particularly limited. The polymerizable functional group is not particularly limited, but each of the one or more polymerizable functional groups included in the reactive terminal structure is, for example, any of an epoxy group, a hydroxy group, an amino group, an isocyanate group, a mercapto group, a vinyl group, a (meth)acrylic group, a carboxyl group, a formyl group, a norbornene ring, a maleimide ring, and an oxazine ring. That is, the bond structure includes, for example, one or more of an epoxy group, a hydroxy group, an amino group, an isocyanate group, a mercapto group, a vinyl group, a (meth)acrylic group, a carboxyl group, a formyl group, a norbornene ring, a maleimide ring, and an oxazine ring. By having one or more polymerizable functional groups in the bond structure shown in the structural information, the bond structure can function as a monomer or oligomer for resin generation. That is, according to the chemical structure proposal device 10, structural information that can be used for resin generation can be efficiently obtained.

[0025] 4 is a diagram illustrating a functional configuration of the chemical structure proposal device 10 according to this embodiment. In the example of FIG. 4, the chemical structure proposal device 10 includes a first group storage unit 101, a second group storage unit 102, a third group storage unit 103, and a bond structure storage unit 104. A reactive terminal bond processor 130 can access each of the first group storage unit 101, the second group storage unit 102, and the third group storage unit 103.

[0026] When the first group storage unit 101 is provided inside the chemical structure proposal device 10, the first group storage unit 101 is realized, for example, by using the storage device 1080 of the computer 1000. When the second group storage unit 102 is provided inside the chemical structure proposal device 10, the second group storage unit 102 is realized, for example, by using the storage device 1080 of the computer 1000. When the third group storage unit 103 is provided inside the chemical structure proposal device 10, the third group storage unit 103 is realized, for example, by using the storage device 1080 of the computer 1000. When the bond structure storage unit 104 is provided inside the chemical structure proposal device 10, the bond structure storage unit 104 is realized, for example, by using the storage device 1080 of the computer 1000. However, one or more of the first group storage unit 101, the second group storage unit 102, the third group storage unit 103, and the bond structure storage unit 104 may be provided outside the chemical structure proposal device 10.

[0027] The first group storage unit 101 holds information about the first group. The information about the first group includes information indicating a plurality of reactive end structures belonging to the first group. As described above, the first group is composed of a plurality of reactive end structures. The information indicating the reactive end structure may be, for example, a descriptor indicating the reactive end structure, or may be identification information such as a symbol defined for each reactive end structure. For example, a database that associates the identification information with the reactive end structure may be separately provided, and the reactive end structure may be identified based on the identification information.

[0028] 5(a) to 5(e) are diagrams illustrating reactive end structures belonging to the first group. The reactive end structure is a part of a molecule. The reactive end structure is a structure having only one bonding point and including a polymerizable functional group. The bonding point can also be said to be an unbonded bond. The bonding points can be bonded to each other. In FIG. 5(a) to FIG. 5(e), the bonding point is a part surrounded by a solid circle, and the polymerizable functional group is a part surrounded by a dashed circle. FIG. 5(a) to FIG. 5(e) show an example in which the polymerizable functional group is an epoxy group. The multiple reactive end structures belonging to the first group are different from each other. The first group is preferably composed of multiple reactive end structures having the same polymerizable functional group. However, one first group may include multiple reactive end structures having different polymerizable functional groups. The reactive end structure may have multiple polymerizable functional groups, but preferably has only one polymerizable functional group.

[0029] When the first group is composed of a plurality of reactive end structures having the same polymerizable functional group, the first group may be determined for each polymerizable functional group. For example, a first group G1e composed of a plurality of reactive end structures having an epoxy group, a first group G1h composed of a plurality of reactive end structures having a hydroxyl group, a first group G1a composed of a plurality of reactive end structures having an amino group, etc. are provided. The first group storage unit 101 can hold information about the first group provided for each of the plurality of polymerizable functional groups. The information about the first group includes information indicating the plurality of reactive end structures belonging to the first group. In this case, the reactive end binding processing unit 130 can select and use any one of the plurality of first groups held in the first group storage unit 101.

[0030] The second group storage unit 102 holds information about the second group. The information about the second group includes information indicating a plurality of main structures belonging to the second group. As described above, the second group is made up of a plurality of main structures. The information indicating the main structure may be, for example, a descriptor indicating the main structure, or may be identification information such as a symbol defined for each main structure. For example, a database that associates the identification information with the main structure may be separately provided, and the main structure may be identified based on the identification information.

[0031] 6(a) to 6(e) are diagrams illustrating main structures belonging to the second group. The main structure is a part of a molecule. The main structure has multiple bonding points. The main structure may or may not have a polymerizable functional group. In FIG. 6(a) to FIG. 6(e), the bonding points are the parts surrounded by solid line circles. The multiple main structures belonging to the second group are different from each other.

[0032] The third group storage unit 103 holds information about the third group. The third group is composed of a plurality of non-reactive end structures. The information about the third group includes information indicating a plurality of non-reactive end structures belonging to the third group. The information indicating the non-reactive end structure may be, for example, a descriptor indicating the non-reactive end structure, or may be identification information such as a symbol defined for each non-reactive end structure. For example, a database that associates the identification information with the non-reactive end structure may be separately provided, and the non-reactive end structure may be identified based on the identification information.

[0033] Figures 7(a) to 7(e) are diagrams illustrating non-reactive end structures belonging to the third group. The non-reactive end structure is a part of a molecule. The non-reactive end structure is a structure that has only one binding point and does not contain a polymerizable functional group. In Figures 7(a) to 7(e), the binding point is the part surrounded by a solid circle.

[0034] The chemical structure proposing device 10 according to the present embodiment generates structural information indicating a bond structure in which at least any main structure and any reactive end structure are bonded to each other. The bond structure is a combination of one or more main structures and one or more reactive end structures, or a combination of one or more main structures, one or more reactive end structures, and one or more non-reactive end structures.

[0035] FIG. 8 is a diagram for explaining the concept of the bond structure according to this embodiment. In FIG. 8, the bond points are the parts surrounded by solid-line circles. The chemical structure proposing device 10 determines the structures to be bonded to each of the multiple bond points of the main structure. Then, the bond structure is determined based on the determination result. In this embodiment, the structures to be bonded to each bond point of the main structure are reactive end structures or non-reactive end structures. The resulting bond structure differs depending on which structure is bonded to each bond point.

[0036] According to the chemical structure proposing device 10 of the present embodiment, information showing a wide variety of bond structures can be obtained. Moreover, the bond structures have reactive terminal structures. Therefore, effective bond structures for resin development can be efficiently proposed.

[0037] The processing performed by the chemical structure proposal device 10 will be described in detail below.

[0038] 9 is a flowchart illustrating the flow of processing performed by the chemical structure proposal device 10 according to the present embodiment. When an operation is performed on the chemical structure proposal device 10 to start processing for generating structural information, first, the reactive terminal binding processing unit 130 acquires condition information (step S21). The condition information includes information indicating conditions used in the reactive terminal binding process. The generated structural information may change depending on the condition information.

[0039] The reactive terminal binding processor 130 may read and acquire condition information that is predetermined and stored in a storage unit accessible from the reactive terminal binding processor 130, or may acquire the condition information by accepting an input from a user. The user can input information to the chemical structure proposal device 10 using an input device connected to the chemical structure proposal device 10 (computer 1000). The user can set the condition information depending on what kind of bond structure he or she wants to obtain. In other words, the user can control the tendency of the bond structure to be obtained by adjusting the condition information.

[0040] The condition information includes, for example, a minimum reactive end number FN. FN is a value indicating the lower limit of the reactive end structures contained in the bond structure. FN is an integer equal to or greater than 1. By including FN in the condition information, it is possible to control the minimum number of reactive end structures contained in the bond structure.

[0041] The condition information also includes, for example, information indicating a probability SP1. The probability SP1 is the probability of binding a non-reactive end structure or a reactive end structure to one binding point.

[0042] The condition information may include a predetermined number TN, which is the number of pieces of structure information to be generated by the generating unit 150. TN is an integer equal to or greater than 1.

[0043] As described above, when a first group is defined for each polymerizable functional group, the condition information may include information indicating any of the polymerizable functional groups. The reactive terminal binding unit 130 can select a first group to be used from among the multiple first groups based on the information indicating the polymerizable functional group.

[0044] The condition information may also include information indicating a first main structure, the first main structure being one main structure selected from the second group.

[0045] However, the condition information does not necessarily have to include these pieces of information. FN, SP1, TN, information indicating the polymerizable functional group, and information indicating the first main structure may each be predetermined in the chemical structure proposing device 10, or may be determined randomly.

[0046] Following step S21, the reactive end ligation processor 130 identifies the first main structure (step S22). If the condition information includes information indicating the first main structure, the reactive end ligation processor 130 can identify the first main information based on the condition information. If the condition information does not include information indicating the first main structure, the reactive end ligation processor 130 may, for example, randomly select one first main structure from the second group.

[0047] Then, the reactive end binding processor 130 performs the reactive end binding process (step S23). At the start of the reactive end binding process, the reactive end binding processor 130 according to this embodiment sets the first main structure as a binding target structure.

[0048] Fig. 10 is a flow chart illustrating the flow of the reactive terminal binding process (step S23). Fig. 10 shows an example in which the reactive terminal structure has only one polymerizable functional group and the main structure has no polymerizable functional group. However, the reactive terminal binding process is not limited to this example.

[0049] When the reactive end binding process starts, the reactive end binding processor 130 identifies one or more unbound binding points (i.e., binding points whose binding destination has not been determined) in the binding target structure (step S101). In this embodiment, the reactive end binding processor 130 identifies multiple binding points in the first main structure as unbound binding points.

[0050] Next, the reactive end binding processor 130 identifies the surplus number N of binding points (step S102). The surplus number N is the number of dangling binding points to which structures other than the reactive end structure can be bound. Specifically, the reactive end binding processor 130 identifies the number CN of dangling binding points identified in step S101. Then, the reactive end binding processor 130 calculates N by subtracting the minimum reactive end number FN from the number CN of dangling binding points. As described above, the chemical structure proposal device 10 can accept input of information indicating FN. That is, FN can be specified by the user. However, FN may be determined in advance.

[0051] In step S103, the reactive end binding processor 130 identifies one target binding point among the unbound binding points. The target binding point may be any of the unbound binding points at that time. The reactive end binding processor 130 can randomly select the target binding point from the unbound binding points. The target binding point can be one of multiple binding points of the main structure included in the binding structure indicated by the structural information.

[0052] In step S104, the reactive end binding processor 130 determines whether the surplus number N is greater than zero (step S104). If N is greater than zero (Yes in step S104), the reactive end binding processor 130 determines whether to bind a reactive end structure to the target binding point (step S105).

[0053] In step S105, the reactive end binding processor 130 can determine whether to bind a reactive end structure or a non-reactive end structure to the target binding point using a predetermined probability SP1. That is, the probability of adding a non-reactive end structure can be adjusted according to the probability SP1. As described above, the chemical structure proposal device 10 can accept input of information indicating the probability SP1. In this way, the probability SP1 can be specified by the user. However, the probability SP1 may be determined in advance.

[0054] For example, the reactive end binding processor 130 binds a non-reactive end structure to the target binding point with a probability of SP1. On the other hand, the reactive end binding processor 130 binds a reactive end structure to the target binding point with a probability of 1-SP1. However, the reactive end binding processor 130 may bind a reactive end structure to the target binding point with a probability of SP1 and a non-reactive end structure to the target binding point with a probability of 1-SP1.

[0055] When the reactive end binding processor 130 determines to bind a non-reactive end structure to the target binding point (No in step S105), the reactive end binding processor 130 identifies a non-reactive end structure to be bound to the target binding point (step S106). For example, the reactive end binding processor 130 can randomly select one non-reactive end structure from the third group stored in the third group storage unit 103. That is, the reactive end binding processor 130 can select a non-reactive end structure to be bound to a main structure included in the binding structure from the third group. In this way, the binding structures are diversified.

[0056] As another example, the condition information may include information indicating the non-reactive end structure to be identified in step S106. In this case, the reactive end binding processor 130 identifies the non-reactive end structure indicated in the condition information as the non-reactive end structure to be bound to the target binding site. The identified non-reactive end structure is determined as the structure to be bound to the target binding site.

[0057] Next, the reactive end binding processor 130 reduces the value of the surplus number N by 1 (step S107).

[0058] Then, the reactive end binding processor 130 determines whether or not there is an unprocessed binding point, that is, a binding point for which the structure to be bound has not been determined, among the unprocessed binding points identified in step S101 (step S109). If there is no unprocessed binding point (No in step S109), the reactive end binding processor 130 ends the reactive end binding process. If there is an unprocessed binding point (Yes in step S109), the process performed by the reactive end binding processor 130 returns to step S103, and one of the unprocessed binding points is identified as a new target binding point.

[0059] When the reactive end binding processor 130 determines in step S105 that a reactive end structure is to be bound to the target binding point (Yes in step S105), the reactive end binding processor 130 identifies a reactive end structure to be bound to the target binding point (step S108). For example, the reactive end binding processor 130 can randomly select one reactive end structure from the first group. As another example, information indicating the reactive end structure to be identified in step S108 may be included in the condition information. In that case, the reactive end binding processor 130 identifies the reactive end structure indicated in the condition information as the reactive end structure to be bound to the target binding point. This identified reactive end structure is determined as the structure to be bound to the target binding point.

[0060] In step S104, when it is determined that the surplus number N is not greater than 0 (No in step S104), the reactive end binding processor 130 also performs the process of step S108.

[0061] Following step S108, the reactive end binding processor 130 executes step S109 described above.

[0062] As described above, the reactive end ligation processing unit 130 can execute the reactive end ligation process.

[0063] Returning to FIG. 9, when the reactive end binding process (step S23) is completed, the generation unit 150 outputs the structural information (step S24). The generation unit 150 generates the structural information based on the result of the decision made by the reactive end binding processing unit 130. Therefore, in the example of FIG. 10, the generation unit 150 can generate structural information indicating a binding structure including a predetermined number of FN or more reactive end structures. In this way, the user can obtain a structure that satisfies the desired conditions.

[0064] For example, the structural information indicates at least a plurality of structures (main structure, reactive terminal structure, etc.) included in the bond structure and a bond relationship between a plurality of bond points of the plurality of structures. The structural information may include a descriptor indicating the bond structure.

[0065] After the reactive end ligation process is completed, the generation unit 150 may generate structural information based on all the items determined by the reactive end ligation processing unit 130, or may generate structural information by updating the information every time the reactive end ligation processing unit 130 makes a decision. In other words, the reactive end ligation process and the generation of structural information by the generation unit 150 may be performed in parallel.

[0066] The destination of the structural information output by the generating unit 150 is not particularly limited. For example, the generating unit 150 outputs the structural information to the bond structure storage unit 104. That is, the generating unit 150 stores the structural information in the bond structure storage unit 104. As another example, the generating unit 150 may display the structural information on a display or the like connected to the chemical structure proposal device 10. In that case, the generating unit 150 may display a diagram showing the bond structure on the display. The generating unit 150 may output the structural information to another device.

[0067] Following step S24, the generation unit 150 determines whether or not a predetermined number TN of structural information has been output (step S25). If the predetermined number TN of structural information has not been output (No in S25), the process returns to step S22, and the reactive terminal bonding processing unit 130 identifies the first main structure again. Then, the generation of structural information is repeated. If the predetermined number TN of structural information has been output (Yes in step S25), the process for generating structural information of the chemical structure proposal device 10 ends. In this way, the generation unit 150 can generate multiple pieces of structural information indicating bond structures. Furthermore, each time the reactive terminal bonding process is performed, structural information is generated with a certain degree of randomness. Therefore, a variety of structures are generated.

[0068] Next, the operation and effect of this embodiment will be described. According to this embodiment, the reactive end binding processor 130 performs a reactive end binding process to determine the reactive end structure selected from the first group as the reactive end structure to be bound to the main structure selected from the second group. Therefore, a polymerizable candidate structure can be efficiently obtained.

[0069] Second Embodiment 11 is a diagram illustrating a functional configuration of a chemical structure proposal device 10 according to a second embodiment. The chemical structure proposal device 10 according to this embodiment is the same as the chemical structure proposal device 10 according to the first embodiment, except for the points described below. Moreover, the chemical structure proposal method according to this embodiment is the same as the chemical structure proposal method according to the first embodiment, except for the points described below.

[0070] The chemical structure proposal device 10 according to this embodiment further includes a main bond processing unit 110. The main bond processing unit 110 performs a main structure bonding process that determines that a first main structure selected from the second group is bonded to a second main structure selected from the second group in a bonded structure. In this way, structural information indicating a polymerizable oligomer can be efficiently obtained.

[0071] The main combination processing unit 110 can access both the first group storage unit 101 and the third group storage unit 103.

[0072] The hardware configuration of a computer that realizes the chemical structure proposal device 10 according to this embodiment is shown, for example, in Fig. 3, similarly to the chemical structure proposal device 10 according to the first embodiment. However, a program module that realizes the function of the main combination processing unit 110 of this embodiment is further stored in the storage device 1080 of the computer 1000 that realizes the chemical structure proposal device 10 according to this embodiment.

[0073] FIG. 12 is a diagram for explaining the concept of the bond structure according to this embodiment. In FIG. 12, the bond points are the parts surrounded by solid-line circles. The chemical structure proposing device 10 determines the structures to be bonded to each of the multiple bond points of the main structure. Then, the bond structure is determined based on the determination result. In this embodiment, the structures to be bonded to each bond point of the main structure are the main structure, a reactive end structure, or a non-reactive end structure. The resulting bond structure differs depending on which structure is bonded to each bond point.

[0074] According to the chemical structure proposal device 10 of this embodiment, information showing a wide variety of bond structures can be obtained, similarly to the chemical structure proposal device 10 of the first embodiment. Moreover, the bond structures have reactive terminal structures. Therefore, bond structures effective for resin development can be efficiently proposed. Furthermore, according to the chemical structure proposal device 10 of this embodiment, a structure in which multiple main structures are connected can be proposed. Therefore, structural information showing oligomers can be efficiently obtained.

[0075] The processing performed by the chemical structure proposal device 10 according to this embodiment will be described in detail below.

[0076] 13 is a flowchart illustrating the flow of processing performed by the chemical structure proposal device 10 according to this embodiment. When an operation is performed on the chemical structure proposal device 10 to start processing for generating structural information, first, the main binding processing unit 110 acquires condition information (step S31). The condition information according to this embodiment includes information indicating the conditions used in the main structure binding process in addition to information indicating the conditions used in the reactive terminal binding process. The generated structural information may change depending on the condition information.

[0077] The main combined processing unit 110 may acquire condition information by reading out the condition information that is determined in advance and stored in a storage unit accessible from the main combined processing unit 110, or may acquire the condition information by receiving an input from the user. The user can set the condition information depending on what type of combined structure he or she wants to obtain. In other words, the user can control the tendency of the combined structure to be obtained by adjusting the condition information.

[0078] In this embodiment, the reactive end binding processing unit 130 may acquire condition information from the main binding processing unit 110, or the reactive end binding processing unit 130 may acquire condition information from a memory unit or the like as described in the first embodiment.

[0079] The main binding processor 110 may obtain only information necessary for the processing performed by the main binding processor 110 from among the condition information. The reactive end binding processor 130 may obtain only information necessary for the processing performed by the reactive end binding processor 130 from among the condition information.

[0080] The condition information according to this embodiment is similar to the condition information according to the first embodiment. However, in this embodiment, the condition information may further include the information described below as information indicating conditions used in the main structure combining process.

[0081] The condition information according to this embodiment may include a predetermined number of iterations MN. MN is a value indicating the number of times the main structure bonding process is performed. MN is an integer equal to or greater than 0. By including MN in the condition information, it is possible to control the tendency of the number of connections of the main structures. Note that by setting MN=0, the chemical structure proposal device 10 according to this embodiment does not perform the main structure bonding process, and can function in the same manner as the chemical structure proposal device 10 according to the first embodiment.

[0082] Moreover, the condition information according to the present embodiment includes, for example, information indicating a probability SP2. The probability SP2 is a probability relating to whether a non-reactive terminal structure or a main structure is bound to one binding point. The probability SP2 may be the same as or different from the probability SP1.

[0083] However, the condition information according to this embodiment does not necessarily have to include these pieces of information. MN and SP2 may be determined in advance in the chemical structure proposing device 10, or may be determined randomly.

[0084] Following step S31, the main merger 110 identifies a first main structure (step S32). If the condition information includes information indicating the first main structure, the main merger 110 can identify the first main information based on the condition information. If the condition information does not include information indicating the first main structure, the main merger 110 may, for example, randomly select one first main structure from the second group.

[0085] Then, the main binding processor 110 judges whether MN is zero (step S33). If MN is zero (Yes in step S33), the main binding processor 110 does not perform the main structure binding process, and the reactive end binding processor 130 performs the reactive end binding process (step S36). In this case, the reactive end binding processor 130 according to this embodiment sets the first main structure identified in step S32 as the binding target structure at the start of the reactive end binding process.

[0086] If MN is not zero (No in step S33), the main connection processor 110 performs a main structure connection process (step S34). At the start of the first main structure connection process, the main connection processor 110 according to this embodiment sets the first main structure identified in step S32 as the connection target structure.

[0087] Fig. 14 is a flow chart illustrating the flow of the main structure binding process (step S34). In the example of Fig. 14, the reactive terminal structure has only one polymerizable functional group, and the main structure has no polymerizable functional group. However, the main structure binding process is not limited to this example.

[0088] When the main structure linking process starts, the main link processor 110 identifies one or more unlinked link points in the link target structures (step S201).

[0089] Next, the main bond processor 110 identifies the surplus number N of bond points (step S202). Specifically, the main bond processor 110 identifies the number CN of dangling bond points identified in step S201. Then, the main bond processor 110 calculates N by subtracting the number FN of the lowest reactive ends from the number CN of dangling bond points. As described above, the chemical structure proposal device 10 can accept input of information indicating FN. That is, FN can be specified by the user. However, FN may be determined in advance.

[0090] In step S203, the main connection processor 110 identifies one target connection point from among the unconnected connection points. The target connection point may be any of the unconnected connection points at that time. The main connection processor 110 can randomly select the target connection point from among the unconnected connection points.

[0091] In step S204, the main link processor 110 determines whether the surplus number N is greater than zero (step S204). If N is greater than zero (Yes in step S204), the main link processor 110 determines whether to link the main structure to the target link point (step S205).

[0092] In step S205, the main binding processor 110 can determine whether to bind the main structure or the non-reactive end structure to the target binding point by using a predetermined probability SP2. That is, the probability of adding the non-reactive end structure can be adjusted according to the probability SP2. As described above, the chemical structure proposal device 10 can accept input of information indicating the probability SP2. In this way, the probability SP2 can be specified by the user. However, the probability SP2 may be determined in advance.

[0093] Specifically, in the first main structure binding process, the main binding processor 110 can determine, using a predetermined probability SP2, whether the second main structure or the non-reactive terminal structure should be bound to one of the multiple binding points of the first main structure in the bound structure. Here, the second main structure may be the same as or different from the first main structure.

[0094] For example, the main binding processor 110 binds a non-reactive end structure to a target binding point with a probability of SP2. On the other hand, the main binding processor 110 binds a reactive end structure to a target binding point with a probability of 1-SP2. However, the main binding processor 110 may bind a main structure to a target binding point with a probability of SP1, and bind a non-reactive end structure to a target binding point with a probability of 1-SP2.

[0095] When the main binding processor 110 determines to bind a non-reactive end structure to the target binding point (No in step S205), the main binding processor 110 identifies a non-reactive end structure to be bound to the target binding point (step S206). For example, the main binding processor 110 can randomly select one non-reactive end structure from the third group stored in the third group storage unit 103. That is, the main binding processor 110 can select a non-reactive end structure to be bound to a main structure included in a binding structure from the third group. In this way, the binding structures are diversified.

[0096] As another example, the condition information may include information indicating the non-reactive end structure to be identified in step S206. In this case, the main binding processor 110 identifies the non-reactive end structure indicated in the condition information as the non-reactive end structure to be bound to the target binding site. The identified non-reactive end structure is determined as the structure to be bound to the target binding site.

[0097] Next, the main combination processor 110 reduces the value of the surplus number N by 1 (step S207).

[0098] Then, the main connection processor 110 determines whether or not there are any unprocessed connection points, i.e., connection points for which a structure to be connected has not been determined, among the unprocessed connection points identified in step S201 (step S209). If there are no unprocessed connection points (No in step S209), the main connection processor 110 ends the main structure connection process. If there are unprocessed connection points (Yes in step S209), the process performed by the main connection processor 110 returns to step S203, and one connection point from the unprocessed connection points is identified as a new target connection point.

[0099] If the main link processor 110 determines in step S205 that the main structure is to be linked to the target bond point (Yes in step S205), the main link processor 110 identifies the main structure to be linked to the target bond point (step S208).

[0100] The main structure attached to the target attachment point may be the same as the first main structure or may be different.

[0101] For example, it may be predetermined that the main structure to be linked to the target bond point is the same as the first main structure. In this case, the main link processor 110 specifies the same structure as the first main structure as the main structure to be linked to the target bond point. In this way, the linked structure becomes a structure in which multiple identical main structures are linked together.

[0102] Alternatively, the main merging processor 110 may randomly select one main structure from the second group. As another example, information indicating the main structure to be identified in step S208 may be included in the condition information. In this case, the main merging processor 110 identifies the main structure indicated in the condition information as the main structure to be linked to the target bond point. The identified main structure is determined as the structure to be linked to the target bond point.

[0103] An upper limit of the number of types of main structures that can be included in a combined structure may be determined in advance. Alternatively, the condition information may include an upper limit of the number of types of main structures that can be included in a combined structure. In this case, the main bond processor 110 selects a main structure within a range in which this upper limit is observed during the repetition of step S208. For example, when the number of types of main structures identified up to that point is less than the upper limit, the main bond processor 110 selects a main structure to be bonded to the target bond point from the second group. On the other hand, when the number of types of main structures identified up to that point is the upper limit, the main bond processor 110 selects a main structure to be bonded to the target bond point from among the main structures identified up to that point.

[0104] In step S204, when it is determined that the surplus number N is not greater than 0 (No in step S204), the main combination processor 110 also performs the process of step S208.

[0105] Following step S208, the main combination processor 110 executes step S209 described above.

[0106] As described above, the main combination processor 110 can execute the main structure processing.

[0107] Returning to FIG. 13, when the main structure binding process (step S34) is completed, the main binding processor 110 judges whether the main structure binding process has been performed MN times (step S35). If the main structure binding process has not been performed MN times (No in step S35), the main binding processor 110 performs the main structure binding process again (step S34). In this way, the main binding processor 110 can repeat the main structure binding process multiple times, making it easier to obtain an oligomer of a desired length as a bound structure. However, in the second or subsequent main structure binding processes, a structure that reflects the results of the determination made by the main binding processor 110 in one or more previous main structure binding processes is used as the binding target structure. That is, in the second or subsequent main structure binding processes, the binding target structure is a structure in which multiple main structures are bound, or a structure in which one or more main structures are bound to one or more non-reactive terminal structures.

[0108] When the main structure binding process has been performed MN times (Yes in step S35), the reactive end binding processor 130 then performs the reactive end binding process (step S36). In this case, the reactive end binding processor 130 according to this embodiment sets the structure that reflects the results of all the decisions made by the main binding processor 110 in all the main structure binding processes performed up to that point as the binding target structure. That is, in the reactive end binding process, the binding target structure is a structure in which multiple main structures are bound, or a structure in which one or more main structures are bound to one or more non-reactive end structures.

[0109] The reactive end binding process (step S36) according to this embodiment is the same as the reactive end binding process (step S23) according to the first embodiment.

[0110] When the reactive end binding process (step S36) is completed, the generation unit 150 outputs the structural information (step S37). The generation unit 150 generates the structural information based on the results of the decisions made by the main binding unit 110 and the reactive end binding unit 130.

[0111] After the reactive end binding process is completed, the generation unit 150 may generate structural information based on all the items determined by the main binding processing unit 110 and the reactive end binding processing unit 130, or may generate structural information by updating information every time the main binding processing unit 110 or the reactive end binding processing unit 130 makes a decision. That is, the main structure binding process and the generation of structural information by the generation unit 150 may be performed in parallel. Also, the reactive end binding process and the generation of structural information by the generation unit 150 may be performed in parallel.

[0112] The output of the structural information is as explained in the first embodiment.

[0113] Following step S37, the generation unit 150 determines whether or not a predetermined number TN of structural information has been output (step S38). If the predetermined number TN of structural information has not been output (No in S38), the process returns to step S32, and the main merger processing unit 110 identifies the first main structure again. Then, the generation of structural information is repeated. If the predetermined number TN of structural information has been output (Yes in step S38), the process for generating structural information of the chemical structure proposal device 10 ends.

[0114] As described above, according to this embodiment, the same actions and effects as those of the first embodiment can be obtained. In addition, according to this embodiment, the main structure binding process is performed by the main binding processor 110. Therefore, the structural information showing the polymerizable oligomer can be efficiently obtained.

[0115] (Third embodiment) 15 is a flowchart illustrating the flow of processing performed by the chemical structure proposal device 10 according to the third embodiment. The chemical structure proposal device 10 according to this embodiment is the same as the chemical structure proposal device 10 according to the second embodiment, except for the points described below. Moreover, the chemical structure proposal method according to this embodiment is the same as the chemical structure proposal method according to the second embodiment, except for the points described below.

[0116] In the chemical structure proposing device 10 according to the present embodiment, the main linking processor 110 determines whether to perform the main structure linking process using a predetermined probability OP. This makes it possible to adjust the linking probability of the main structures and increases the variety of generated structure information.

[0117] In this embodiment, the condition information includes, for example, information indicating the probability OP. That is, the main merger processing unit 110 can accept input of information indicating the probability OP. Therefore, the probability OP can be specified by the user. However, the condition information does not need to include information indicating the probability OP. In the chemical structure proposal device 10, the probability OP may be predetermined or may be randomly determined.

[0118] The process performed by the chemical structure proposal device 10 according to this embodiment will be described in detail below with reference to Fig. 15. Steps S41 to S43 are the same as steps S31 to S33 described in the second embodiment. If MN is not zero in step S33 (No in step S33), the main binding processor 110 performs step S44 before performing the main structure binding process.

[0119] In step S44, the main connection processor 110 judges whether or not to perform the main structure connection process. For example, the main connection processor 110 next performs the main structure connection process (step S45) with a probability of OP (Yes in step S44). On the other hand, the main connection processor 110 next performs step S46 without performing the main connection process with a probability of 1-OP (No in step S44). However, the main connection processor 110 may next perform step S46 without performing the main connection process with a probability of OP, and next perform the main structure connection process (step S45) with a probability of 1-OP.

[0120] The main structure combining process (step S45) according to this embodiment is the same as the main structure combining process (step S34) according to the second embodiment. The main combining processor 110 according to this embodiment performs step S46 following the main structure combining process (step S45).

[0121] In step S46, the main binding processor 110 judges whether or not the above-mentioned step S44 has been performed MN times. If step S44 has not been performed MN times (No in step S46), the main binding processor 110 performs step S44 again. If step S44 has been performed MN times (Yes in step S46), the reactive end binding processor 130 then performs reactive end binding processing (step S47).

[0122] The reactive end binding process (step S47) according to this embodiment is the same as the reactive end binding process (step S36) according to the second embodiment. Moreover, steps S48 and S49 are the same as steps S37 and S38, respectively, described in the second embodiment.

[0123] As described above, according to this embodiment, the same actions and effects as those of the second embodiment can be obtained. In addition, according to this embodiment, the main connection processing unit 110 determines whether to perform the main structure connection process using a predetermined probability OP. This makes it possible to adjust the connection probability of the main structure and increases the variety of generated structure information.

[0124] (Fourth embodiment) 16 is a diagram illustrating a functional configuration of a chemical structure proposal device 10 according to a fourth embodiment. The chemical structure proposal device 10 according to this embodiment is the same as the chemical structure proposal device 10 according to at least one of the first to third embodiments, except for the points described below. Moreover, the chemical structure proposal method according to this embodiment is the same as the chemical structure proposal method according to at least one of the first to third embodiments, except for the points described below.

[0125] FIG. 16 shows an example in which the chemical structure proposal apparatus 10 is equipped with a main bond processing unit 110, similar to the chemical structure proposal apparatus 10 according to the second and third embodiments. However, the chemical structure proposal apparatus 10 according to this embodiment may not be equipped with the main bond processing unit 110, similar to the chemical structure proposal apparatus 10 according to the first embodiment.

[0126] In this embodiment, the chemical structure proposal device 10 further includes an estimation unit 170. The estimation unit 170 estimates the characteristics of the bond structure using a machine learning model, for example. The estimated characteristics are not particularly limited, but include, for example, one or more of specific gravity, flexural modulus, tensile modulus, flexural strength, tensile strength, elongation, impact strength, adhesion strength, volume resistivity, refractive index, relative dielectric constant, dielectric loss tangent, thermal conductivity, linear expansion coefficient, water absorption rate, melting point, and cure shrinkage rate. The estimated characteristics (referred to as "bond structure characteristics") may be characteristics of the substance itself represented by the bond structure, or may be characteristics of a resin obtained under specified conditions using the substance represented by the bond structure. The estimated characteristics may also be characteristics of a cured body or composition obtained under specified conditions using the resin. The estimated characteristics may be, for example, the specific gravity of a cured body of a resin obtained using the substance represented by the bond structure.

[0127] The hardware configuration of a computer that realizes the chemical structure proposal device 10 according to this embodiment is, for example, shown in Fig. 3, similar to the chemical structure proposal device 10 according to the first embodiment. However, a program module that realizes the function of the estimation unit 170 of this embodiment is further stored in the storage device 1080 of the computer 1000 that realizes the chemical structure proposal device 10 according to this embodiment.

[0128] The estimation unit 170 estimates the characteristics of the bond structure, for example, by using a trained model that has been previously trained in machine learning. This machine learning is performed, for example, by using training data including information indicating a chemical structure and characteristics (correct answer data) related to the chemical structure. The chemical structure in the training data is preferably a structure having a polymerizable functional group. In addition, the chemical structure in the training data preferably includes the same polymerizable functional group as the bond structure. The characteristics related to the chemical structure in the multiple training data used in the machine learning are preferably unified to specific conditions such as measurement conditions, resin formation conditions (reactivity, etc.), and curing conditions. In this way, characteristics based on the specific conditions are estimated as the characteristics of the bond structure.

[0129] The trained model used by the estimation unit 170 includes a neural network. Structural information indicating a bond structure can be input to the trained model. In addition, the information output from the trained model includes one or more pieces of information indicating characteristics.

[0130] The estimation unit 170 may estimate only one characteristic, or may estimate multiple characteristics (for example, specific gravity, bending modulus, and tensile modulus, etc.). When the chemical structure proposal device 10 estimates multiple characteristics, the estimation unit 170 may use a trained model capable of outputting multiple pieces of information indicating the characteristics, or may use multiple trained models capable of outputting only one piece of information indicating the characteristic.

[0131] FIG. 17 is a flowchart illustrating the flow of processing performed by the chemical structure proposal device 10 according to this embodiment. In this example, steps S51 to S58 are the same as steps S41 to S48 described in the third embodiment, respectively. In step S58 of this embodiment, the generation unit 150 outputs structural information to the estimation unit 170. When the estimation unit 170 acquires structural information from the generation unit 150, it estimates the characteristics of the bond structure indicated by the structural information (step S59). That is, the estimation unit 170 inputs the structural information into the trained model. Then, information indicating the characteristics of the bond structure is obtained as an output of the trained model. The estimation unit 170 outputs the structural information in association with the information indicating the characteristics of the bond structure.

[0132] There is no particular limitation on the destination to which the estimation unit 170 outputs the information. For example, the estimation unit 170 outputs the structural information and the information indicating the characteristics to the bond structure storage unit 104. That is, the estimation unit 170 causes the bond structure storage unit 104 to hold the structural information and the information indicating the characteristics. As another example, the estimation unit 170 may display the structural information and the information indicating the characteristics on a display or the like connected to the chemical structure proposal device 10. In that case, the estimation unit 170 may display a diagram showing the bond structure on the display, or may display a graph or chart showing the characteristics. The estimation unit 170 may output the structural information and the information indicating the characteristics to another device.

[0133] Step S60 is the same as step S49 in the third embodiment.

[0134] 17, an example has been described in which the chemical structure proposal device 10 performs the same process as the chemical structure proposal device 10 according to the third embodiment and further estimates the properties. However, the chemical structure proposal device 10 may perform the same process as the chemical structure proposal device 10 according to the first or second embodiment and further estimate the properties.

[0135] Furthermore, the estimation unit 170 may further estimate the synthesis difficulty of the bond structure in addition to or instead of estimating the characteristics of the bond structure. The estimation unit 170 may also use a trained model on which machine learning has been performed in advance for estimating the synthesis difficulty. This machine learning is performed, for example, using training data including information indicating a chemical structure and the synthesis difficulty (correct answer data) of a substance having the chemical structure. Structural information indicating a bond structure can be input to this trained model. Furthermore, the information output from the trained model includes the synthesis difficulty.

[0136] As described above, according to this embodiment, the same actions and effects as those of at least one of the first to third embodiments can be obtained. In addition, according to this embodiment, the chemical structure proposal device 10 further includes an estimation unit 170. This allows additional information to be obtained for the bond structure, and allows material development to proceed efficiently. EXAMPLES

[0137] The present embodiment will be described in detail below with reference to examples. However, the present embodiment is not limited to the description of these examples.

[0138] Example 1 Using the chemical structure proposal method described in the first embodiment, multiple pieces of structural information were generated for the purpose of searching for resins with low specific gravity. In generating the structural information, the minimum number of reactive terminals FN was set to 2, and the probability SP1 (probability of non-reactive terminal addition) was set to 0.6. The polymerizable functional group was a hydroxyl group. All of the bond structures indicated by the structural information obtained in this example were polymerizable structures. Furthermore, for each piece of obtained structural information, the specific gravity was estimated using a trained model, and the bond structure indicated by the structural information with the lowest estimated specific gravity was designated as bond structure 1.

[0139] Example 2 A plurality of pieces of structural information were generated using the chemical structure proposal method described in the third embodiment. The number of pieces of structural information generated was the same as in Example 1. In generating the structural information, the minimum number of reactive terminals FN was set to 2, the number of repetitions MN was set to 2, the probability OP (probability of executing the main structure bonding process) was set to 0.5, the probability SP1 (probability of adding a non-reactive terminal) was set to 0.6, and the probability SP2 (probability of adding a non-reactive terminal) was set to 0.6. The polymerizable functional group was set to a hydroxyl group. All of the bond structures indicated by the structural information obtained in this example were polymerizable structures. Furthermore, for each piece of structural information obtained, the specific gravity was estimated using a trained model, and the bond structure indicated by the structural information with the lowest estimated specific gravity was set to bond structure 2.

[0140] (Comparative Example) A plurality of pieces of structural information were generated without using the information of the first group as described in the embodiment. That is, in this comparative example, a reactive end binding process was not performed to determine a reactive end structure selected from the first group consisting of a plurality of reactive end structures as a reactive end structure to be bound to the main structure. In generating the structural information, the number of repetitions MN was set to 0, the probability OP was set to 0.5, and the probability SP2 was set to 0.6. All of the binding structures indicated by the structural information obtained in this comparative example were structures that could not be polymerized. That is, the structural information of the comparative example was a structure that could not generate a resin, and was not useful information in resin development. An example of a binding structure indicated by the structural information generated in the comparative example is called binding structure 3.

[0141] Figures 18(a) to 18(c) are diagrams showing bond structure 1 to bond structure 3, respectively. Bond structure 2 was an oligomer. On the other hand, bond structure 1 and bond structure 3 were not oligomers.

[0142] As described below, the substances represented by bonding structure 1 and bonding structure 2 were polymerized to produce resins, and the specific gravity of the resulting resins was measured. As described above, bonding structure 3 cannot produce resins, so the specific gravity could not be measured.

[0143] To generate the resin, we first prepared substance A having the structure shown in Figure 18(a), substance B having the structure shown in Figure 19(a), substance C having the structure shown in Figure 19(b), and substance D having the structure shown in Figure 19(c). Substances A and B were synthesized independently. The results of identifying substances A and B by 1H NMR are shown in Figures 20 and 21, respectively. As substance C, we used epoxy resin 1:4-(2,3-epoxypropan-1-yloxy)-N,N-bis(2,3-epoxypropan-1-yl)-2-methylaniline (EMA) manufactured by Osaka Soda Co., Ltd. As substance D, we used TPP (Tetraphenylporphyrin) manufactured by Tokyo Chemical Industry Co., Ltd.

[0144] In Example 1, 76.9 wt % of substance A, 22.3 wt % of substance C, and 0.8 wt % of substance D were mixed and cured at 180° C. for 2 hours to obtain a cured resin body 1.

[0145] In Example 2, 57.4 wt% of substance A, 33.4 wt% of substance B, 8.3 wt% of substance C, and 0.9 wt% of substance D were mixed and cured at 180° C. for 2 hours to obtain a resin cured product 2. By mixing and curing substance A and substance B, an oligomer corresponding to bond structure 2 was formed in the mixed system.

[0146] The specific gravities of the hardened body 1 and the hardened body 2 were measured by Archimedes' method. The specific gravity of the hardened body 1 was 1.06, and the specific gravity of the hardened body 2 was 1.04. In this way, in both Example 1 and Example 2, a resin with a desired low specific gravity could be efficiently searched for. Furthermore, in Example 2, a bond structure (oligomer) that can obtain a resin with a lower specific gravity could be searched for compared to Example 1.

[0147] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various configurations other than those described above can also be adopted. [Explanation of symbols]

[0148] 10 Chemical structure proposal device 101 First group memory unit 102 Second group memory unit 103 Third Group Memory Unit 104 Connection structure storage 110 Main connection processing section 130 Reactive end ligation processing section 150 Generation part 170 Estimation Department 1000 calculator 1020 Bus 1040 Processor 1060 Memory 1080 Storage Device 1100 Input / Output Interface 1120 Network Interface

Claims

1. One or more computers performing a reactive end binding process to determine the reactive end structure selected from a first group consisting of a plurality of reactive end structures as the reactive end structure to be bound to the main structure selected from a second group consisting of a plurality of main structures; generating information indicating a bond structure obtained by bonding the main structure selected from the second group and the reactive terminal structure selected from the first group; the main structure has a plurality of attachment points; The reactive end structure has only one attachment point and includes a polymerizable functional group. Chemical structure proposal method.

2. The chemical structure proposing method according to claim 1, the one or more computers determine, using a predetermined probability SP1, whether the reactive end structure or the non-reactive end structure is to be bonded to one of a plurality of bonding points of the main structure included in the bonded structure; The non-reactive terminal structure has only one attachment point and does not contain a polymerizable functional group. Chemical structure proposal method.

3. The chemical structure proposing method according to claim 2, The one or more computers accept input of information indicating the probability SP1. Chemical structure proposal method.

4. The chemical structure proposing method according to claim 2 or 3, The one or more computers select the non-reactive terminal structure to be bound to the main structure included in the binding structure from a third group consisting of a plurality of the non-reactive terminal structures. Chemical structure proposal method.

5. The chemical structure proposing method according to claim 1 or 2, The one or more computers generate information indicating the bond structure including a predetermined number FN or more of the reactive terminal structures. Chemical structure proposal method.

6. The chemical structure proposing method according to claim 5, The one or more computers accept input of information indicating the number FN. Chemical structure proposal method.

7. The chemical structure proposing method according to claim 1, The one or more computers further perform a main structure combining process for determining, in the combined structure, combining a first main structure selected from the second group with a second main structure selected from the second group. Chemical structure proposal method.

8. The chemical structure proposing method according to claim 7, the one or more computers determine, using a predetermined probability SP2, whether the second main structure or the non-reactive terminal structure is to be bonded to one of a plurality of bonding points of the first main structure in the bonded structure; The non-reactive terminal structure has only one attachment point and does not contain a polymerizable functional group. Chemical structure proposal method.

9. The chemical structure proposing method according to claim 8, The one or more computers accept input of information indicating the probability SP2. Chemical structure proposal method.

10. The chemical structure proposing method according to claim 7 or 8, The one or more computers repeat the main structure binding process multiple times. Chemical structure proposal method.

11. The chemical structure proposing method according to claim 7 or 8, The one or more computers determine whether to perform the main structure joining process using a predetermined probability OP. Chemical structure proposal method.

12. The chemical structure proposing method according to claim 11, The one or more computers accept input of information indicating the probability OP. Chemical structure proposal method.

13. The chemical structure proposing method according to claim 1 or 2, The polymerizable functional group is any one of an epoxy group, a hydroxy group, an amino group, an isocyanate group, a mercapto group, a vinyl group, a (meth)acrylic group, a carboxyl group, a formyl group, a norbornene ring, a maleimide ring, and an oxazine ring. Chemical structure proposal method.

14. The chemical structure proposing method according to claim 1 or 2, The one or more computers generate a plurality of pieces of information indicating the bonding structure. Chemical structure proposal method.

15. The chemical structure proposing method according to claim 1 or 2, The one or more computers further estimate properties of the bond structure using a machine learning model. Chemical structure proposal method.

16. The chemical structure proposing method according to claim 15, The characteristic is the specific gravity of the cured resin obtained by using the substance having the bonding structure. Chemical structure proposal method.

17. The chemical structure proposal method according to claim 1 or 2, the first group is defined for each polymerizable functional group, The one or more computers further comprise: Obtaining condition information including information indicating any one of the polymerizable functional groups; A first group to be used is selected from the plurality of first groups based on the information indicating the polymerizable functional group. Chemical structure proposal method.

18. A program that causes a computer to function as a chemical structure proposal device, The computer a reactive end linking processing means for performing a reactive end linking processing to determine the reactive end structure selected from a first group consisting of a plurality of reactive end structures as the reactive end structure to be linked to the main structure selected from a second group consisting of a plurality of main structures; a generating means for generating information indicating a bond structure obtained by bonding the main structure selected from the second group and the reactive terminal structure selected from the first group; It functions as the main structure has a plurality of attachment points; The reactive end structure has only one attachment point and includes a polymerizable functional group. program.

19. a reactive end binding processing unit that performs a reactive end binding process to determine the reactive end structure selected from a first group consisting of a plurality of reactive end structures as the reactive end structure to be bound to the main structure selected from a second group consisting of a plurality of main structures; a generating unit that generates information indicating a bond structure obtained by bonding the main structure selected from the second group and the reactive terminal structure selected from the first group; Equipped with the main structure has a plurality of attachment points; The reactive end structure has only one attachment point and includes a polymerizable functional group. Chemical structure proposal device.

Images