Complex molecular structure automatic generation system, method, and program

Abstract

To provide a complex molecular structure automatic generation system, method, and program with which it is possible to generate the molecular structure of a complex molecule easily with high accuracy.SOLUTION: The complex molecular structure automatic generation system according to an embodiment comprises an initial arrangement unit, a structure optimization unit, a distance reduction unit, and a successive processing unit. The initial arrangement unit, regarding a complex molecule having a central atom and a plurality of ligands coordinated around the central atom, arranges a plurality of ligands at initial positions away from the central atom by an initial distance. The structure optimization unit optimizes the structure of the plurality of ligands. The distance reduction unit reduces the distance between the central atom and the plurality of ligands following a prescribed rule. The successive processing unit successively executes optimization by the structure optimization unit and distance reduction by the distance reduction unit until a prescribed termination condition is satisfied.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] An embodiment of the present invention relates to a system, method, and program for automatically generating a complex molecular structure. [Background technology]

[0002] Bright red phosphors with excellent color purity are required for display applications. Micro LED displays, which are attracting attention as a new generation of display, require red phosphors that are not only bright and have high color purity, but also have small particle sizes to eliminate unevenness in the emitted color. However, with the currently mainstream inorganic red phosphors, the luminescence quantum yield decreases as the particle size decreases.

[0003] One solution to this problem is organic fluorescent materials, or fluorescent complexes, which have molecular-based luminescence and become transparent when dissolved. Regarding fluorescent complexes, Eu(III) complexes have been the focus of research to date, focusing on red-emitting complexes, and efforts to improve their luminescence intensity and solubility have focused on them. Eu(III) ions alone exhibit low light absorption and weak luminescence, but ligands such as β-diketones absorb light and transfer energy to the Eu(III) ions, increasing their luminescence intensity. The efficiency of energy transfer is known to be highly dependent on the R1 and R2 substituents of the β-diketone ligands. Furthermore, in addition to ionic ligands such as β-diketones, nonionic ligands such as phosphine oxides have also been considered as promising candidates for rare earth complexes. In fact, it has been shown that the luminescence intensity can be further enhanced by coordinating phosphine oxides to the Eu(III) ion in addition to β-diketones. Furthermore, in recent years, it has been reported that the luminescence intensity can be further increased and the solubility in polymers can be improved by using two different types of phosphine oxides coordinated to one Eu(III) rather than using multiple phosphine oxides with the same structure.

[0004] In this way, the design of red fluorescent complexes involves many non-trivial combinations of ligand arrangements around the metal atom and molecular structures, making the search for candidate complex molecules enormous. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] H. Iwanaga. “Investigation of strong photoluminescence and highly soluble eu (iii) complexes with phosphine oxides and β-diketonates”. Journal of Luminescence, Vol. 200, pp. 233-239, 2018. Summary of the Invention [Problem to be solved by the invention]

[0006] The problem to be solved by the present invention is to provide a system, method and program for automatically generating a molecular structure of a complex, which is capable of generating the molecular structure of a complex molecule easily and with high accuracy. [Means for solving the problem]

[0007] An automatic complex molecular structure generation system according to an embodiment includes an initial placement unit, a structure optimization unit, a distance reduction unit, and a sequential processing unit. The initial placement unit, for a complex molecule having a central atom and multiple ligands coordinated around the central atom, places the multiple ligands at initial positions that are an initial distance away from the central atom. The structure optimization unit optimizes the structures of the multiple ligands. The distance reduction unit reduces the distance between the central atom and the multiple ligands in accordance with a predetermined rule. The sequential processing unit sequentially performs optimization by the structure optimization unit and distance reduction by the distance reduction unit until a predetermined termination condition is satisfied. [Brief explanation of the drawings]

[0008] [Figure 1] A diagram showing an example of the hardware configuration of the automatic complex molecular structure generation system [Figure 2] A diagram showing an example of the functional configuration of a processor [Figure 3] Diagram showing an example of the molecular structure of a complex molecule [Figure 4] A diagram showing the procedure for automatically generating the molecular structure of a complex molecule. [Figure 5] Diagram showing the molecular structure of an eight-coordinate Eu(III) complex [Figure 6] A schematic diagram showing the preprocessing procedure for the placement operation of a bonding atom when there is one bonding atom. [Figure 7] A schematic diagram showing the procedure for placing a bonding atom when there is one bonding atom. [Figure 8] A schematic diagram showing the procedure for placing bonding atoms when there are two bonding atoms. [Figure 9] FIG. 1 is a diagram schematically illustrating sequential processing of structure optimization by a structure optimization unit and distance reduction by a distance reduction unit. [Figure 10] A diagram showing an example of a display screen of the most stable structure [Figure 11] FIG. 10 is a diagram illustrating a process for generating a characteristic parameter map. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, a system, method, and program for automatically generating a complex molecular structure according to this embodiment will be described with reference to the drawings.

[0010] Fig. 1 is a diagram showing an example of the hardware configuration of an automatic complex molecular structure generation system 100. As shown in Fig. 1, the automatic complex molecular structure generation system 100 is a computer having a processor 1, a read-only memory (ROM) 2, a random access memory (RAM) 3, an auxiliary storage device 4, an input device 5, a display device 6, and a communication device 7. Transmission and reception of data and various signals between the processor 1, ROM 2, RAM 3, auxiliary storage device 4, input device 5, display device 6, and communication device 7 is performed via a bus.

[0011] The processor 1 is an integrated circuit that controls the overall operation of the complex molecular structure automatic generation system 100. For example, the processor 1 has a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), and / or a floating-point unit (FPU). The processor 1 may also have an internal memory and an I / O interface. The processor 1 executes various processes by interpreting and calculating programs stored in advance in a ROM 2, an auxiliary storage device 4, or the like. The processor 1 may also be implemented in part or in whole by hardware such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

[0012] The ROM 2 is a non-volatile memory that stores various data. For example, the ROM 2 stores data and setting values ​​used when the processor 1 executes various processes. The ROM 2 may include a non-transitory computer-readable storage medium that stores programs executed by the processor 1.

[0013] The RAM 3 is a volatile memory used for reading and writing data. The RAM 3 temporarily stores data used when the processor 1 executes various processes. The RAM 3 provides a work area for the processor 1.

[0014] The auxiliary storage device 4 is a non-volatile memory that stores various data. For example, the auxiliary storage device 4 stores data and setting values ​​used when the processor 1 executes various processes, data generated by the various processes in the processor 1, etc. The auxiliary storage device 4 is configured by an HDD (Hard Disk Drive), an SSD (Solid State Drive), an integrated circuit storage device, etc. The auxiliary storage device 4 may also include a non-transitory computer-readable storage medium that stores a program executed by the processor 1.

[0015] The input device 5 accepts various operation inputs from an operator. Examples of the input device 5 that can be used include a keyboard, a mouse, various switches, a touchpad, and a touch panel display. An electrical signal corresponding to the accepted operation input (hereinafter referred to as an operation signal) is supplied to the processor 1.

[0016] The display device 6 displays various data under the control of the processor 1. A CRT (Cathode-Ray Tube) display, a liquid crystal display, an organic EL (Electro Luminescence) display, an LED (Light-Emitting Diode) display, a plasma display, or any other display may be used as the display device 6. The display device 6 may also be a projector.

[0017] The communication device 7 includes a communication interface such as a network interface card (NIC) for communicating data with various devices connected to the complex molecular structure automatic generation system 100 via a network. Note that an operation signal may be supplied from a computer connected via the communication device 7 or an input device included in the computer, and various data may be displayed on a display device or the like included in the computer connected via the communication device 7. However, for the sake of simplicity of the following description, unless otherwise specified, it is assumed that the source of the operation signal is the input device 5 and the display destination of the various data is the display device 6. The input device 5 can be replaced by a computer connected via the communication device 7 or an input device included in the computer, and the display device 6 can be replaced by a display device or the like included in the computer connected via the communication device 7.

[0018] The automatic complex molecular structure generation system 100 does not need to include all of the ROM 2, RAM 3, auxiliary storage device 4, input device 5, display device 6, and communication device 7. If necessary, some of the ROM 2, RAM 3, auxiliary storage device 4, input device 5, display device 6, and communication device 7 may be omitted. The automatic complex molecular structure generation system 100 may also include any additional hardware device useful for executing the processes according to this embodiment. The automatic complex molecular structure generation system 100 does not need to be physically composed of a single computer, but may be composed of a computer system having multiple computers communicably connected via wires, a network, or the like. The series of processes according to this embodiment can be assigned to multiple processors 1 implemented on multiple computers, respectively, in any manner. All processors 1 may execute all processes in parallel, or specific processes may be assigned to one or some of the processors 1, and the series of processes according to this embodiment may be executed by the entire computer system.

[0019] Fig. 2 is a diagram showing an example of the functional configuration of the processor 1. As shown in Fig. 2, the processor 1 has functional components such as an input unit 11, an initial placement unit 12, a structure optimization unit 13, a distance reduction unit 14, a sequential processing unit 15, a steric collision determination unit 16, a quantum chemistry calculation unit 17, a map generation unit 18, and a display control unit 19. The processor 1 automatically generates the molecular structure of a complex molecule through cooperation of all or part of these functional components.

[0020] FIG. 3 is a diagram showing an example of the molecular structure of a complex molecule 30. As shown in FIG. 3, the complex molecule 30 has a central atom 31 and a plurality of ligands 32 coordinated around the central atom 31. The central atom 31 is the atom at the center of the complex molecule 30. The central atom 31 is assumed to be a metal atom, but may be a non-metal atom. The ligand 32 is a molecule or ion that bonds to the central atom 31. The ligand 32 has a plurality of atoms, and among these atoms, an atom that is directly bonded to the central atom 31 is called a bonding atom 33. There may be one bonding atom 33 per ligand 32, or there may be two or more bonding atoms 33.

[0021] The input unit 11 inputs structural condition expressions and various processing conditions of the complex molecule. The structural condition expressions are series data of letters, symbols, and / or numbers that represent structural conditions such as the type of central atom, the coordination number of multiple ligands, the molecular structure of the multiple ligands, and / or the coordination coordinates of the multiple ligands. The coordination number means the number of ligands coordinated to the central atom. The coordination coordinates mean the coordinates at which the ligands can coordinate. The molecular structure represents the type of molecules contained in the ligand and the arrangement of the molecules. The coordination coordinates are determined according to the coordination structure of the complex molecule. The various processing conditions input include the reduction distance and the termination condition of the sequential processing, which are used for automatically generating the molecular structure of the complex molecule.

[0022] The initial placement unit 12 places a plurality of ligands at an initial position away from the central atom by an initial distance for a complex molecule having a central atom and a plurality of ligands coordinated around the central atom. Hereinafter, the distance between the central atom and the ligands is referred to as the central atom-ligand distance. The initial distance is at least twice the distance between the central atom and the plurality of ligands. Hereinafter, the first distance refers to the central atom-ligand distance of a complex molecule expected to be structurally stable. Specifically, the initial placement unit 12 reconstructs a three-dimensional molecular structure of the complex molecule based on the structural condition expression input by the input unit 11, and places a bonding atom bonded to the central atom of each of the plurality of ligands included in the three-dimensional molecular structure at one coordinate (initial position) among a plurality of coordination coordinates that can be coordinated in the general structure of the complex molecule.

[0023] The structure optimization unit 13 optimizes the structures of multiple ligands. The multiple ligands optimized by the structure optimization unit 13 include multiple ligands arranged by the initial arrangement unit 12 and multiple ligands after the central atom-ligand distance has been reduced by the distance reduction unit 14. The structure optimization unit 13 fixes only the bond-related portion of each of the multiple ligands and optimizes the structure of the other portions. The structure optimization unit 13 optimizes the structures of the multiple ligands by displacing the coordinates of the components of each of the multiple ligands based on the force acting on the component.

[0024] The distance reduction unit 14 reduces the distance between the central atom and the plurality of ligands (central atom-ligand distance) according to a predetermined rule. As the predetermined rule, the distance reduction unit 14 reduces the central atom-ligand distance by a predetermined ratio or a predetermined distance. The predetermined ratio or the predetermined distance can be input via the input unit 11, for example.

[0025] The sequential processing unit 15 sequentially executes optimization by the structure optimization unit 13 and reduction of the central atom-ligand distance by the distance reduction unit 14 until a predetermined termination condition is satisfied. When it is determined that the predetermined termination condition is satisfied, the sequential processing unit 15 outputs the most stable structure, which is the most stable structure among the structures optimized each time the central atom-ligand distance is reduced up to the time when the determination was made.

[0026] The steric clash determination unit 16 determines whether or not there is a collision for all or some combinations of the multiple atoms constituting the central atom and the multiple ligands. The determination of whether or not there is a collision by the steric clash determination unit 16 is not essential. The steric clash determination unit 16 determines whether or not there is a collision at any stage in the process of sequentially performing optimization by the structure optimization unit 13 and reduction of the central atom-ligand distance by the distance reduction unit 14. If the steric clash determination unit 16 determines that there is a collision, the initial placement unit 12 places multiple ligands at other initial positions, and then the sequential processing unit 15 sequentially performs optimization by the structure optimization unit 13 and distance reduction by the distance reduction unit 14 until a predetermined termination condition is satisfied.

[0027] The quantum chemistry calculation unit 17 performs quantum chemistry calculations on the most stable structure output by the sequential processing unit 15 .

[0028] The map generating unit 18 generates a map in which characteristic parameters of a plurality of most stable structures corresponding to a plurality of complex molecules are plotted in an N-dimensional space defined by N (N is a natural number of 1 or more) structural parameters related to the structures.

[0029] The display control unit 19 displays various information on the display device 6. As one example, the display control unit 19 displays the most stable structure of the complex molecule output by the sequential processing unit 15. As another example, the display control unit 19 displays the map generated by the map generation unit 18. As another example, the display control unit 19 may display the calculation results of the quantum chemistry calculation performed by the quantum chemistry calculation unit 17.

[0030] Next, an example of the operation of the complex molecular structure automatic generation system 100 will be described.

[0031] 4 is a diagram showing the processing procedure for automatic generation of the molecular structure of a complex molecule. As shown in FIG. 4, the input unit 11 inputs a structural condition expression of a complex molecule (step S1). In step S1, the input unit 11 inputs, as the structural condition expression, series data of letters, symbols, and / or numbers representing structural conditions such as the type of central atom, the coordination number of a ligand, the molecular structure of a ligand, and / or the coordination coordinates of a ligand. Each element of the structural condition expression can be determined arbitrarily.

[0032] Hereinafter, the complex molecule is referred to as a metal complex molecule. That is, the central atom is a metal atom. The type of metal atom may be any metal atom, such as manganese, aluminum, iron, nickel, copper, silver, gold, europium, or the like. The coordination number, molecular structure, and coordination coordinates are also not particularly limited.

[0033] After step S1 is performed, the initial placement unit 12 restores the molecular structure of the complex molecule from the structural condition expression input in step S1 (step S2). In step S2, the initial placement unit 12 restores the three-dimensional molecular structure of the complex molecule based on the structural condition expression. Specifically, the initial placement unit 12 restores the three-dimensional molecular structure of each of the central atoms and each of the ligands constituting the complex molecule individually. The restoration of the molecular structure of the ligand may be performed by a well-known first method (G. Landrum. Rdkit documentation. Release, Vol. 1, No. 1-79, pp. 4, 2013) or a second method (N.M. O'Boyle, M. Banck, C.A. James, C. Morley, T. Vandermeersch, and G.R. Hutchison. Open babel: An open chemical toolbox. Journal of cheminformatics, Vol. 3, No. 1, pp. 1-14, 2011), etc.

[0034] To make molecular structures, which can be complex three-dimensional, easier to handle as data, graph-based molecular graph representation methods are often used. The idea behind molecular graph representation is to map the atoms and bonds that make up a molecule to a set of nodes and edges. The atoms that make up a molecule are treated as nodes, and the bonds as edges, and the atomic arrangement and connections between atoms are converted into data structures such as matrices or arrays. A typical application of molecular graph representation is SMILES (Simplified Molecular Input Line Entry System) (D. Weininger. Smiles, a chemical language and information system. 1. Introduction to methodology and encoding rules. Journal of chemical information and computer sciences, Vol. 28, No. 1, pp. 31-36, 1988). SMILES representation is an example of the above structural condition representation.

[0035] In the SMILES representation, a number is assigned to each atom in a complex molecule, and by scanning the molecular graph in that order, a unique character string representation of the molecular structure can be obtained. The SMILES representation allows molecular structures to be easily handled as data even on a computer. This embodiment also uses molecular information given in the SMILES representation. However, since the SMILES representation is a conversion of the three-dimensional structure of a molecule into a character string representation and does not contain information about the three-dimensional structure itself, it is generally difficult to reversely restore the three-dimensional structure from the character string information in the SMILES representation. A method for converting a metal complex molecule into a character string representation using the SMILES representation has not yet been established.

[0036] Therefore, in this embodiment, the expression of a complex molecule is separated into a central metal and a ligand, and the three-dimensional molecular structure of the ligand is reconstructed from the SMILES expression. The entire metal complex molecule is reconstructed by performing coordination search, structure optimization, distance reduction, and steric clash determination, which will be described later. As a result, one complex molecule can be represented by a single list of SMILES expressions of the ligands. By providing a group of ligand lists, a comprehensive range of metal complexes can be handled as data, and the three-dimensional molecular structure of each complex can be reproduced in this embodiment.

[0037] After step S2 is performed, the input unit 11 inputs various processing conditions such as a reduction distance and a termination condition (step S3). The reduction distance is a parameter related to the distance for reducing the central atom-ligand distance, which is used in step S8. The termination condition is a parameter related to the termination condition of the sequential processing, which is used in step S10.

[0038] After step S3 is performed, the initial placement unit 12 searches for the coordination coordinates of the bonding atom (step S4). In step S4, the initial placement unit 12 determines the coordinates for arranging the bonding atom bonded to the central atom of each of the plurality of ligands included in the three-dimensional molecular structure from among a plurality of candidate coordination coordinates that can be coordinated in the general structure of the complex molecule.

[0039] Step S4 will be explained below using an eight-coordinate metal complex molecule of a europium (Eu(III)) complex as a specific example. In the case of an eight-coordinate metal complex molecule of a Eu(III) complex, the central atom is Eu(III) and the coordination number is eight. As an example, the eight-coordinate molecular structure is assumed to be a square antiprism (sqap). The coordination candidate coordinates can be determined in advance depending on the molecular structure of the ligand.

[0040] Figure 5 shows the molecular structure of an eight-coordinate Eu(III) complex. As shown in Figure 5, in an eight-coordinate Eu(III) complex, Eu(III) is positioned as the central atom, and eight ligands are positioned around Eu(III). The coordination candidate coordinate sqap_coord for sqap can be given as a three-dimensional coordinate as follows:

[0041] sqap_coord = [[-1.67465, -1.67465, -0.44628], #0 [-1.67465, 1.67465, 0.44628], #1 [-0.89035, -0.89035, 2.05491], #2 [-0.89035, 0.89035, -2.05491], #3 [1.67465, -1.67465, 0.44628], #4 [1.67465, 1.67465, -0.44628], #5 [0.89035, -0.89035, -2.05491], #6 [0.89035, 0.89035, 2.05491]] #7

[0042] First, the initial placement unit 12 determines a molecule (bonding atom) that bonds to Eu(III) from among the multiple ligand molecules contained in the three-dimensional molecular structure. The bonding atom is determined from among the multiple atoms that make up the ligand molecule and have an unshared electron pair. Next, the initial placement unit 12 determines the coordination coordinates of the bonding atom.

[0043] For example, if the ligand molecule is a β-diketone or phosphine oxide, the oxygen atom of the unshared electron pair in the ligand is determined as the bonding atom. Then, for the oxygen atom determined as the bonding atom, any of the coordinates 0 to 7 of the above sqap_coord is determined as the coordination coordinate. The coordination coordinate is determined at a random or predetermined position from any of the coordinates 0 to 7.

[0044] More specifically, in the case of a molecule containing two oxygen atoms coordinated to a metal atom, such as a β-diketone, the initial placement unit 12 determines in advance pairs of candidate coordination coordinates from sqap_coord. Hereinafter, pairs of candidate coordination coordinates are referred to as coordinate pairs. In the case of Figure 5, coordinates #0 and #2 are coordinate pairs. Similarly, coordinates #1 and #3, #4 and #6, and #5 and #7 are coordinate pairs. The initial placement unit 12 randomly selects one coordinate pair from multiple coordinate pairs, and determines one coordinate of the selected coordinate pair as the coordination coordinate of one oxygen atom, and the other coordinate as the coordination coordinate of the other oxygen atom.

[0045] In the case of a molecule such as phosphine oxide, in which there is one oxygen atom coordinated in the molecule, the initial placement unit 12 determines in advance the coordinate candidate pairs from sqap_coord, randomly selects one pair from among the multiple coordinate pairs, and determines one of the selected coordinate pairs as the coordination coordinate of the oxygen atom. Note that when determining this coordination coordinate, care is taken to ensure that it does not overlap with the coordination coordinate of another oxygen atom.

[0046] The selection of coordinate pairs is not limited to random selection, but may be systematic or user-selected. The bonding atom is not limited to an oxygen atom, but any atom contained in the ligand may be selected. While the molecular structure is described as sqap, it is not limited to this, and any type of molecular structure may be determined.

[0047] After step S4 is performed, the initial placement unit 12 places the ligand based on the coordination coordinates determined in step S4 (step S5). More specifically, in step S5, the initial placement unit 12 places the bonding atom of the ligand in the complex molecule restored in step S2 at the coordination coordinates searched for in step S4. This placement operation is performed for each ligand. The procedure for the placement operation will be explained separately for the cases where one bonding atom is included in one ligand and the case where two bonding atoms are included. First, the case where there is one bonding atom will be explained.

[0048] FIG. 6 is a diagram showing a schematic diagram of the preprocessing procedure for the arrangement operation of the bonding atom when there is one bonding atom. In step S2, the three-dimensional molecular structure of the complex molecule is restored, and the three-dimensional molecular structure of the ligand 32 of the complex molecule is shown in the left diagram of FIG. 6. The ligand 32 contains one bonding atom 33. The type of bonding atom 33 is not particularly limited, but it is assumed to be an oxygen atom. It is assumed that the three-dimensional molecular structure of the complex molecule is restored in an xyz Cartesian coordinate system.

[0049] As shown in Figure 6, the initial placement unit 12 first moves the bonding atom 33 to the origin of the xyz Cartesian coordinate system. Next, the initial placement unit 12 rotates the vector 35 connecting the bonding atom 33 and its adjacent atom 34 by an angle Θ so that the vector 35 becomes parallel to the z-axis of the three-dimensional coordinate system. This preprocessing makes it possible to move the bonding atom 33 in the ligand 32 closer to the desired coordination coordinates with a simple operation. The initial placement unit 12 performs this preprocessing for each ligand. Note that this preprocessing can be omitted as appropriate.

[0050] After the above preprocessing, the initial placement unit 12 places the ligand by placing the bonding atoms at the coordination coordinates determined in step S4 while maintaining the three-dimensional molecular structure of the ligand. Note that the coordinate information of each atom in the xyz Cartesian coordinate system is represented as normalized_xyz. This embodiment has the advantage that the three-dimensional coordinate data normalized_xyz can be obtained from the SMILES representation.

[0051] Figure 7 is a diagram showing the procedure for placing a bonding atom when there is one bonding atom. In Figure 7, the bonding atom is an oxygen atom and the central atom is Eu(III). In this case, the distance between the bonding atom and the central atom is expressed as dis_o_eu.

[0052] When there is one bonding atom 33, a rotation operation is performed from normalized_xyz obtained in preprocessing to any one of the coordination coordinates in spqa_coords. Hereinafter, any one of the coordination coordinates in spqa_coords will be referred to as Eu_coord. The initial placement unit 12 sets the central metal Eu(III) at the coordinate origin, places the ligand 32 at a coordinate space a distance dis_o_eu away in the Z-axis direction, and rotates the ligand 32 by θ1 around the origin, thereby placing the bonding atom 33 at Eu_coord. The rotation angle θ1 can be calculated based on the dot product of the z-axis and the vector Eu_coord.

[0053] The initial placement unit 12 then moves each ligand along the line connecting the central atom and each bonding atom so that the distance between the central atom and the bonding atom becomes the initial distance. The initial distance is set to a distance at least twice the expected stable distance. The expected stable distance is set to the central atom-ligand distance of a complex molecule that is expected to be structurally stable. This completes the placement of the ligand 32.

[0054] Next, the procedure for arranging the ligands when there are two bonding atoms will be explained.

[0055] First, the initial placement unit 12 performs preprocessing for each ligand. When there are two bonding atoms, the initial placement unit 12 specifies one of the two bonding atoms and performs the preprocessing shown in FIG. 6 for the specified bonding atom. Any method may be used to specify the bonding atom. After the preprocessing, the initial placement unit 12 places the two bonding atoms in the ligand at the two coordination coordinates determined in step S4 while maintaining the three-dimensional molecular structure of the ligand. At this time, placement optimization is performed so that the two bonding atoms are as close as possible to the two coordination coordinates while maintaining the three-dimensional molecular structure of the ligand.

[0056] FIG. 8 is a diagram showing the procedure for placing bonding atoms when there are two bonding atoms. When there are two bonding atoms in coordination, the number of Eu_coords is two. In this case, first, the placement of bonding atom 331 to one of the two Eu_coords (hereinafter referred to as Eu_coord1) is performed using the same procedure as when there is one bonding atom, and then the placement of bonding atom 332 to the other Eu_coord (hereinafter referred to as Eu_coord2) is performed using a two-stage optimization. This is explained in detail below. In the preprocessing, bonding atom 331 corresponds to bonding atom 33 in FIG. 6, and therefore bonding atom 331 is assumed to be located on the Z axis.

[0057] First, the initial placement unit 12 rotates the first bonding atom 331 by a rotation angle θ2 and places it at Eu_coord1 while maintaining the three-dimensional molecular structure of the ligand 32. Next, to place the other bonding atom 332 at Eu_coord2, the initial placement unit 12 rotates the ligand 32 by a rotation angle θ3 around a line 37 connecting the origin and Eu_coord1 as the rotation axis so that bonding atom 332 is on a two-dimensional plane 36 spanned by a vector from the origin passing through Eu_coord1 and a vector from the origin passing through Eu_coord2.

[0058] Next, in order to bring bonding atom 332 on two-dimensional plane 36 closer to Eu_coord2, initial placement unit 12 calculates angle θ4 between vector 38 connecting Eu_coord1 and bonding atom 332 and vector 39 connecting Eu_coord1-Eu_coord2 from the dot product of these two vectors, and rotates ligand 32 by angle θ4 around vector 40 perpendicular to two-dimensional plane 36 as the rotation axis. Because it is usually difficult to precisely align bonding atom 332 with Eu_coord2, initial placement unit 12 finely increments rotation angle θ4, determines the rotation angle that minimizes the distance between Eu_coord2 and bonding atom 332, and rotates ligand 32 by that rotation angle to optimize the placement.

[0059] Then, the initial placement unit 12 moves each ligand along the line connecting the central atom and each bonding atom so that the distance between the central atom and the bonding atom becomes the above initial distance. As a result, the ligand 32 can be placed based on Eu-coord1 and Eu-coord2.

[0060] After step S5 is performed, the structure optimization unit 13 optimizes the molecular structure of the ligand placed in step S5 (step S6). In step S6, the structure optimization unit 13 determines the optimal position of all atoms in all ligands, excluding the central atom, based on the force acting on each atom. As an example, the structure optimization unit 13 calculates the sum of the forces acting on each atom while displacing the position of each atom, and determines the optimal position of each atom so that the sum is minimized. As an example, the force acting on each atom is calculated based on the potential energy in molecular dynamics. This optimizes the molecular structure of the ligand. However, the positions of the bonding atoms in the ligand are fixed. For structure optimization, a representative semi-empirical force field, MMFF94 (Merck Molecular Force 94), can be used. It is also possible to specify other general-purpose force fields, such as MM (Molecular Mechanics) 3, MM4, AMBER (Assisted Model Building and Energy Refinement), and CHARMM (Chemistry at Harvard Macromolecular Mechanics).

[0061] After step S6 is performed, the sequential processing unit 15 determines whether or not the central atom-ligand distance needs to be reduced (step S7). In step S7, the sequential processing unit 15 determines whether or not a reduction end condition is satisfied. The reduction end condition may be set, for example, that the central atom-ligand distance has reached a lower limit, that the number of times the reduction process has been executed has reached an upper limit, or the like. If the reduction end condition is not satisfied, it is determined that reduction is required, and if the reduction end condition is satisfied, it is determined that reduction is not required.

[0062] If it is determined in step S7 that the central atom-ligand distance needs to be reduced (step S7: YES), the distance reduction unit 14 reduces the central atom-ligand distance (step S8). In step S8, the sequential processing unit 15 reduces the distance between the central atom and the plurality of ligands in accordance with a predetermined rule each time the structural optimization in step S4 is completed. As the predetermined rule, the central atom-ligand distance is reduced by a predetermined ratio or a predetermined distance. For example, the predetermined ratio may be set to a value greater than 0 and less than 1 with respect to the central atom-ligand distance before reduction. This makes it possible to finely divide the central atom-ligand distance.

[0063] 9 is a diagram schematically illustrating the sequential processing of structure optimization by the structure optimization unit 13 and distance reduction by the distance reduction unit 14. As shown in Fig. 9, after reducing the central atom-ligand distance, the sequential processing unit 15 sequentially performs structure optimization by the structure optimization unit 13 in step S6 and distance reduction by the distance reduction unit 14 in step S8 for the reduced central atom-ligand distance until it is determined in step S7 that the central atom-ligand distance should not be reduced.

[0064] If it is determined in step S7 that the central atom-ligand distance does not need to be reduced (step S7: NO), the steric clash determination unit 16 determines whether or not there is a steric clash with respect to the most stable structure (step S9). In step S9, the steric clash determination unit 16 first outputs the most stable molecular structure (hereinafter referred to as the most stable structure) among the molecular structures optimized each time the central atom-ligand distance is reduced up to the time the determination was made. The most stable structure corresponds to, for example, the molecular structure with the lowest energy among the molecular structures of a plurality of complex molecules corresponding to each of a plurality of central atom-ligand distances.

[0065] Next, the steric clash determination unit 16 determines whether or not there is a steric clash with respect to the most stable structure. In step S5, it is possible to arrange the ligands at the desired coordination coordinates. However, if the ligands are arranged in an unnatural manner during the process of optimizing the arrangement in step S5, collisions between the ligands may occur. In generating a metal complex molecular structure, collisions between the coordinates of different atoms do not cause any abnormalities, but if this structure is executed as input for quantum chemical calculations, there is a risk of abnormal termination. To avoid this, the steric clash determination unit 16 determines whether or not there is a steric clash in order to determine whether or not the most stable structure is correct.

[0066] Specifically, first, the steric clash determination unit 16 defines the minimum interatomic distance d_min. d_min can be set to any value, for example, 1.0 angstrom. Next, the steric clash determination unit 16 calculates the distances between all atoms constituting the complex molecule, and if any one of the interatomic distances is below d_min, it determines that a steric clash exists. The determination of the presence or absence of a steric clash is made based on the xyz coordinate data of the complex molecule. Note that this determination is merely an example, and it may also be determined that a steric clash exists when all or two or more of all interatomic distances are below d_min. Alternatively, d_min may be compared with a statistical value such as the average of all or some of the interatomic distances.

[0067] When step S9 is performed, the 3D collision determination unit 16 determines whether the termination condition for the sequential processing is satisfied (step S10). The termination condition for the sequential processing is set, for example, when the number of determinations for the presence or absence of a 3D collision by the 3D collision determination unit 16 reaches an upper limit, or when the 3D collision determination unit 16 determines that there is no collision. The termination condition in FIG. 4 is assumed to be when the 3D collision determination unit 16 determines that there is no collision.

[0068] If it is determined in step S10 that the termination condition for the sequential processing is not satisfied (step S10: NO), the sequential processing unit 15 discards the current most stable structure, returns to step S5, and instructs the initial placement unit 12, structure optimization unit 13, distance reduction unit 14, and steric clash determination unit 16 to perform steps S5 to S9 again. The re-optimization of the configuration in step S5 is preferably performed so as to obtain a different configuration from the previous configuration optimization in step S5. That is, the multiple ligands constituting the complex molecule are placed at different initial positions.

[0069] If it is determined in step S10 that the termination condition for the sequential processing is satisfied (step S10: YES), the sequential processing unit 15 outputs the most stable structure of the complex molecule at the current time (step S11). As an example, the most stable structure may be output as three-dimensional coordinate data for each atom constituting the complex molecule, or as image data graphically representing the complex molecule. According to this embodiment, the most stable structure can be easily and accurately obtained for a large-scale complex molecule by sequentially performing structure optimization and distance reduction. Furthermore, it is also possible to obtain the most stable structure that avoids steric clashes by performing steric clash determination.

[0070] When step S11 is performed, the display control unit 19 displays the most stable structure output in step S11 on the display device 6 (step S12). The most stable structure may be displayed in any layout.

[0071] FIG. 10 is a diagram showing an example of a display screen I1 of the most stable structure. As shown in FIG. 10, the display screen I1 has display fields I11, I12, and I13. Display field I11 displays image data that graphically represents the most stable structure of the complex molecule output in step S11. Displaying the image data allows the user to graphically grasp the most stable structure. Display field I12 displays a numerical value representing the distance between the central metal and the ligand of the most stable structure (central metal-ligand distance), and display field I13 displays a numerical value representing the energy of the most stable structure. This allows the user to grasp the central metal-ligand distance and energy of the most stable structure.

[0072] The layout of the display screen I1 is not limited to the layout shown in Fig. 10, and any items may be displayed. For example, the reduction distance used in step S7 or the termination condition used in step S10 may be displayed. Furthermore, other physical property information apart from the energy may be displayed. If it is not necessary to display the most stable structure, step S12 can be omitted.

[0073] When step S12 is performed, the process of automatically generating the molecular structure of the complex molecule shown in FIG. 4 is completed.

[0074] The most stable structure is then subjected to quantum chemistry calculations, for example, by the quantum chemistry calculation unit 17. The quantum chemistry calculation unit 17 performs quantum chemistry calculations using the most stable structure as an initial structure to obtain various physical property information such as energy, energy transfer, molecular structure, charge distribution, molecular orbital, and chemical reaction. The quantum chemistry calculation method is not particularly limited, and any method such as the Hartree-Fock method, molecular orbital method, density functional theory, semi-empirical method, or the like may be used. The quantum chemistry calculation unit 17 may be implemented by a classical computer or a quantum computer. According to this embodiment, as described above, a highly accurate most stable structure can be obtained. Therefore, by using such a highly accurate most stable structure as the initial structure for the quantum chemistry calculation, it is possible to complete the quantum chemistry calculation while avoiding errors in the quantum chemistry calculation.

[0075] The above-described automatic molecular structure generation process for complex molecules allows for the automatic generation of a three-dimensional molecular structure for each complex molecule from a string list of ligands expressed in SMILES format. The reconstructed three-dimensional molecular structure can be used as input for quantum chemical calculations, enabling the theoretical prediction of the properties of a given complex molecule with high accuracy. Because the method for generating complex molecules can be automated on a computer, comprehensive automatic calculations can be performed using complex molecule data. As a result, the properties of candidate molecules can be obtained automatically without experimentation, significantly reducing the development time and costs required for the vast number of molecular combinations.

[0076] The processing procedure for automatically generating the molecular structure of a complex molecule shown in FIG. 4 is an example, and the present embodiment is not limited to this, and various elements can be deleted, added, and / or changed.

[0077] As an example, the determination of the presence or absence of a steric clash (S9) may not be performed. In this case, the termination condition in step S10 may be set to the fact that the number of times the iterative processing of steps S5 to S10 has been executed reaches an upper limit, or may be set unconditionally (step S10 may be omitted).

[0078] As another example, the determination of the presence or absence of a steric clash (S9) does not necessarily have to be performed every time it is determined that distance reduction will not be performed (S7: NO), and may be performed at other times. Specifically, it may be performed every time distance reduction (S8) is performed, or every time structure optimization (S6) is performed. Alternatively, the determination of the presence or absence of a steric clash (S9) may be performed every several times steps S6 to S8 are repeated.

[0079] The input of various processing conditions (S3) may be performed at different times depending on the input item. Specifically, the input of the reduction distance may be performed before step S7, and the input of the termination condition for the sequential processing may be performed before step S10. If the various processing conditions are input automatically, the input of various processing conditions (S3) can be omitted.

[0080] The termination condition for step S10 is not limited to one, and multiple termination conditions may be set. For example, in step S10, a first termination condition may be determined as to whether or not the number of times the presence or absence of a steric collision has been determined has reached an upper limit. If it is determined that the upper limit has not been reached, a second termination condition may be determined as to whether or not a steric collision has occurred. If it is determined that a steric collision has occurred, the process may return to step S5. If it is determined that the number of times the presence or absence of a steric collision has been determined has reached an upper limit or if it is determined that no steric collision has occurred, the process may proceed to step S11.

[0081] If it is determined in step S10 that the termination condition is not satisfied, the process returns to step S5, but this embodiment is not limited to this, and the process may return to any of steps S1, S2, S3, and S4. For example, by returning to step S1, it becomes possible to repeatedly perform steps S1 to S10 by changing the type of central atom, ligand, etc., molecular structure, etc.

[0082] Here, we will explain an example in which a europium complex is used as the complex molecule. A europium atom is set as the central atom, and β-diketone and phosphine oxide are set as the ligands. The bonding atom in the ligand is an oxygen atom. When performing sequential structural optimization of the ligand, the interatomic distance between the europium atom and the oxygen atom is set sufficiently large in the initial structure so that the ligand molecules do not collide. In this example, dis_o_eu was set to 4.6 angstroms. The SMILES representations of the three β-diketones and phosphine oxides are as follows:

[0083] smi0 = 'O=C([CH]C(=O)(C(F)(F)F))C(F)(F)C(F)(F)C(F)(F)F' smi4_0 = 'P(=O)c1c(cccc1)Oc1c(P(=O))cccc1'

[0084] The interatomic distance between the europium atom and the oxygen atom was reduced by a ratio of 0.9 per optimization, and this process was repeated eight times to obtain multiple most stable structures. It was confirmed that all of them had an eight-coordinate Eu(III) structure and no steric clashes occurred. Furthermore, it was confirmed that some of these structures could be used as initial structures for quantum chemical calculations, and that quantum chemical calculations could be performed without errors.

[0085] 4 is executed while changing the structural conditions, it is possible to obtain a plurality of most stable structures corresponding to a plurality of complex molecules having different structural conditions. The map generation unit 18 generates a map (hereinafter referred to as a characteristic parameter map) in which the characteristic parameters of the plurality of most stable structures are plotted in an N-dimensional space defined by N (N is a natural number equal to or greater than 1) structural parameters related to the structure.

[0086] Fig. 11 is a diagram schematically illustrating the process of generating a characteristic parameter map. As shown in the left diagram of Fig. 11, the map generating unit 18 inputs a plurality of most stable structures corresponding to a plurality of complex molecules having different structural conditions. The structural conditions include the type of central atom, the coordination number of the ligand, the molecular structure of the ligand, the coordination coordinates of the ligand, and / or the arrangement coordinates of the bonding atoms.

[0087] The map generation unit 18 determines N structural parameter values ​​from each of the multiple most stable structures. The type of structural parameter may be selected from among the elements of the structural conditions, such as the type of central atom, the coordination number of the ligand, the molecular structure of the ligand, the coordination coordinates of the ligand, and / or the location coordinates of the bonding atoms. Alternatively, a feature vector output by a machine learning model may be used as the structural parameter, or any other structural parameter may be calculated. The machine learning model is trained to input three-dimensional coordinate data and / or image data of the most stable structure and output N structural parameters. For example, the machine learning model may be the encoder portion of an autoencoder that sequentially encodes and decodes the three-dimensional coordinate data and / or image data of the most stable structure.

[0088] Furthermore, the map generating unit 18 calculates a characteristic parameter value for each of the multiple most stable structures. Any type of parameter that represents the characteristics of the most stable structure, such as energy or energy transfer, may be used as the type of characteristic parameter value. The map generating unit 18 generates a characteristic parameter map by plotting the characteristic parameter values ​​in N-dimensional space.

[0089] The right diagram in Figure 11 is an example of a characteristic parameter map for N=2. The horizontal axis represents the first structural parameter (PC1), the vertical axis represents the second structural parameter (PC2), and the characteristic parameter is energy transfer. An energy transfer value, which is a characteristic parameter, is assigned to each pixel in the two-dimensional space formed by PC1 and PC2. This makes it possible to generate a characteristic parameter map that represents a two-dimensional distribution of energy transfer values. Note that for pixels that do not have the most stable structure, it is preferable to estimate the energy transfer value based on the energy transfer values ​​of neighboring pixels that have the most stable structure.

[0090] The display control unit 19 displays the generated characteristic parameter map on the display device 6. Each pixel in the characteristic parameter map is displayed with a color value corresponding to the assigned energy transfer value. The display control unit 19 may also display marks 41 at noteworthy points such as extreme values ​​in the two-dimensional distribution. Displaying the characteristic parameter map makes it possible to visually grasp the two-dimensional distribution of characteristic parameter values ​​such as energy transfer values, and is expected to provide design guidelines for new complex molecules.

[0091] As in the above embodiment, the complex molecular structure automatic generation system 100 has an initial placement unit 12, a structure optimization unit 13, a distance reduction unit 14, and a sequential processing unit 15. The initial placement unit 12 places multiple ligands at initial positions that are an initial distance away from the central atom for a complex molecule having a central atom and multiple ligands coordinated around the central atom. The structure optimization unit 13 optimizes the structure of the multiple ligands. The distance reduction unit 14 reduces the distance between the central atom and the multiple ligands in accordance with a predetermined rule. The sequential processing unit sequentially performs optimization by the structure optimization unit and distance reduction by the distance reduction unit until a predetermined termination condition is met.

[0092] According to the above configuration, the structural optimization of a complex molecule is divided into the optimization of the ligand structure and the reduction of the central atom-ligand distance, and both are performed sequentially. According to the method of this embodiment, the risk of steric clashes occurring is reduced compared to when the determination of the central atom-ligand distance is also performed during structural optimization, and therefore it is possible to easily optimize the molecular structure of a complex molecule with high accuracy.

[0093] Thus, according to this embodiment, it is possible to provide a system, method, and program for automatically generating a complex molecular structure that can generate the molecular structure of a complex molecule easily and with high accuracy.

[0094] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0095] 1...processor, 2...ROM, 3...RAM, 4...auxiliary storage device, 5...input device, 6...display device, 7...communication device, 11...input section, 12...initial placement section, 13...structure optimization section, 14...distance reduction section, 15...sequential processing section, 16...steric collision judgment section, 17...quantum chemistry calculation section, 18...map generation section, 19...display control section, 30...complex molecule, 31...central atom, 32...ligand, 33...bonding atom, 34...adjacent atom, 100...complex molecular structure automatic generation system.

Claims

1. an initial placement unit that places the plurality of ligands at initial positions spaced an initial distance from the central atom in a complex molecule having a central atom and a plurality of ligands coordinated around the central atom; a structure optimization unit that optimizes the structures of the plurality of ligands; a distance reduction unit that reduces the distance between the central atom and the plurality of ligands according to a predetermined rule; a sequential processing unit that sequentially executes the optimization by the structural optimization unit and the distance reduction by the distance reduction unit until a predetermined termination condition is satisfied; An automatic complex molecular structure generation system comprising:

2. 2. The automatic complex molecular structure generation system according to claim 1, wherein the structure optimization unit fixes only a portion of each of the plurality of ligands that is involved in bonding, and optimizes the structure of the remaining portion.

3. 2. The automatic complex molecular structure generation system according to claim 1, wherein the structure optimization unit optimizes the structures of the plurality of ligands by displacing the coordinates of the constituent elements of each of the plurality of ligands based on the forces acting on the constituent elements.

4. the initial distance is at least twice as large as a first distance between the central atom and the plurality of ligands; the first distance is a distance between the central atom and the plurality of ligands in the complex molecule that is expected to be structurally stable; The system for automatically generating a complex molecular structure according to claim 1.

5. 2. The system for automatically generating a complex molecular structure according to claim 1, wherein the distance reduction unit reduces the distance between the central atom and the plurality of ligands by a predetermined ratio or a predetermined distance as the predetermined rule.

6. 6. The automatic complex molecular structure generation system according to claim 5, further comprising an input unit for inputting the predetermined ratio or the predetermined distance.

7. a collision determination unit that determines whether or not a collision occurs for all or some combinations of a plurality of atoms that constitute the central atom and the plurality of ligands, the initial placement unit places the plurality of ligands at other initial positions when the collision determination unit determines that a collision has occurred; the sequential processing unit sequentially executes the optimization by the structural optimization unit and the distance reduction by the distance reduction unit until the predetermined termination condition is satisfied. The system for automatically generating a complex molecular structure according to claim 1.

8. 8. The automatic complex molecular structure generation system according to claim 7, wherein the predetermined termination condition is that the number of times the collision determination unit determines whether or not a collision exists reaches an upper limit or that the collision determination unit determines that no collision exists.

9. 8. The automatic complex molecular structure generation system according to claim 7, wherein the collision determination unit determines whether or not a collision occurs at any stage in the process of sequentially performing optimization by the structure optimization unit and distance reduction by the distance reduction unit.

10. 2. The system for automatically generating a complex molecular structure according to claim 1, further comprising an input unit for inputting the predetermined termination condition.

11. 2. The automatic complex molecular structure generation system according to claim 1, wherein, when it is determined that the predetermined termination condition is satisfied, the sequential processing unit outputs a most stable structure that is the most stable structure among the structures optimized each time the distance between the central atom and the plurality of ligands is reduced up to the time when the determination is made.

12. 12. The automatic complex molecular structure generation system according to claim 11, further comprising a quantum chemical calculation unit that performs quantum chemical calculations on the most stable structure.

13. 12. The automatic complex molecular structure generation system according to claim 11, further comprising a map generation unit that generates a map in which characteristic parameters of a plurality of most stable structures corresponding to a plurality of complex molecules are plotted in an N-dimensional space defined by N (N is a natural number of 1 or more) structural parameters related to the structures.

14. 14. The automatic complex molecular structure generation system according to claim 13, further comprising a display control unit that displays the map on a display device.

15. an input unit for inputting a structural condition expression, which is series data of characters, symbols, and / or numerical values ​​representing the type of the central atom, the coordination number of the plurality of ligands, the molecular structures of the plurality of ligands, and / or coordination candidate coordinates of the plurality of ligands; the initial placement unit restores a three-dimensional molecular structure of the complex molecule based on the structural condition expression, and places a bonding atom that is bonded to the central atom of each of the plurality of ligands included in the three-dimensional molecular structure at one coordinate that is the initial position among a plurality of coordination candidate coordinates that can be coordinated in a general structure of the complex molecule. The system for automatically generating a complex molecular structure according to claim 1.

16. the central atom is a metal atom, The complex molecule is a metal complex molecule. The system for automatically generating a complex molecular structure according to claim 1.

17. For a complex molecule having a central atom and a plurality of ligands coordinated around the central atom, the plurality of ligands are arranged at initial positions spaced an initial distance from the central atom; optimizing the structures of the plurality of ligands; reducing the distance between the central atom and the plurality of ligands according to a predetermined rule; optimizing the structures of the plurality of ligands and reducing the distance between the central atom and the plurality of ligands sequentially until a predetermined termination condition is satisfied; A method for automatically generating a complex molecular structure comprising:

18. The processor a function of arranging, for a complex molecule having a central atom and a plurality of ligands coordinated around the central atom, the plurality of ligands at initial positions spaced an initial distance from the central atom; optimizing the structures of the plurality of ligands; a function of reducing the distance between the central atom and the plurality of ligands according to a predetermined rule; a function of sequentially optimizing the structures of the plurality of ligands and reducing the distances between the central atom and the plurality of ligands until a predetermined termination condition is satisfied; This is an automatic molecular structure generation program that realizes the above.

Images