Solubility Evaluation Apparatus, Solubility Evaluation Method, and Solubility Evaluation Program

Abstract

To appropriately evaluate a molecular simulation result to initial conformation and suppress variation in calculation results of solubility features due to differences in the initial conformation.SOLUTION: An initial conformation setting section 11 sets a plurality of initial conformations. A solubility feature is calculated for each molecular simulation executed for each initial conformation. A partial three-dimensional structural information acquisition section 14 calculates partial three-dimensional structural information on a molecule in each molecular simulation. A partial structure identification section 15 identifies a partial structure corresponding to variation in the solubility features on the basis of a solubility feature and partial three-dimensional structural information for each molecular simulation. A variation evaluation section 17 selects a group on the basis of thermodynamic stability of each group in which the solubility feature is classified on the basis of the partial structure, and calculates variation in the solubility features. A variation determination section 18 causes a plurality of new initial conformations to be set when the variation is equal to or greater than a target variation value. An output section 19 outputs evaluation results when the variation is less than the target variation value.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a solubility evaluation apparatus, a solubility evaluation method, and a solubility evaluation program.

Background Art

[0002] In general, in molecular simulations commonly used in molecular dynamics, since the results depend on the initial structure and initial conformation of molecules, different simulation results may be obtained due to differences in the initial structure and initial arrangement. One of the reasons for this is that the local stable states of the system equilibrated by the initial conformation are different, so it is conceivable that the steric structures and energy states sampled during the simulation vary depending on the initial structure. Therefore, when presenting the calculation results by molecular simulation, it is preferable to present the simulation results while also considering the variation in results due to the initial conformation.

[0003] Also, in the molecular dynamics method, the initial conformation dependence of the calculation can be reduced by methods such as the heat annealing method and other methods for transitioning to a more stable local stable state, or by techniques such as the replica exchange method for improving the sampling efficiency. On the other hand, even when these techniques are introduced, evaluating the variation in results for different initial conformations in the final calculation results is still important for indicating the reliability of the calculation results, the robustness of the calculation model, and the procedure.

[0004] For example, as a simulation technique in computational chemistry, a simulation method related to the cleavage of molecular chains of polymer materials has been proposed (for example, Patent Document 1). In addition, a physical property value calculation device having a function of calculating stress data, relaxation modulus, and viscoelasticity from time-series data of molecular dynamics calculations using a polymer model has been proposed (for example, Patent Document 2). These are techniques for evaluating physical property values using the results of molecular simulations. Further, a technique for improving the accuracy of heat conduction calculations using the time average value and time fluctuation of molecular dynamics simulation results has been proposed (for example, Patent Document 3). Furthermore, a technique for evaluating and classifying the calculation results of molecular dynamics for drugs quickly and with high accuracy has been proposed (for example, Patent Document 4). In this technique, the analysis of the results of molecular dynamics is made efficient by training the fluctuation information of the three-dimensional structure of molecules (RMSD: Root Mean Square Deviation) and the labeled calculation results using SVM (Support Vector Machine).

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, with conventional molecular simulation techniques, it is difficult to appropriately evaluate the initial conformation and suppress the variation in the calculation results of solubility characteristics due to differences in the initial conformation. For example, in the technique of evaluating physical property values using the results of the molecular simulation described above, the initial conformation dependence of the calculation results is not considered. Also, the technique of improving the accuracy of physical property value calculation by utilizing the temporal fluctuation (temporal variation) of molecular dynamics simulation is a method based on a single simulation, and the initial conformation dependence is likewise not considered.

[0007] In addition, in the case of a technique that classifies calculation results as favorable or unfavorable using the fluctuation information (RMSD) of the three-dimensional structure of a molecule as an index, it is difficult to appropriately consider the influence of the initial conformation on the calculation results of the solubility characteristics in the classification using RMSD that indicates the similarity of the entire molecule. Therefore, even when using this technique, it is difficult to appropriately evaluate the initial conformation and suppress the variation in the calculation results due to differences in the initial conformation.

[0008] One aspect of the present invention appropriately evaluates the initial conformation and suppresses the variation in the calculation results of solubility characteristics due to differences in the initial conformation.

Means for Solving the Problems

[0009] The solubility evaluation apparatus according to one aspect has the following components. The initial conformation setting unit sets a plurality of different initial conformations. The molecular simulation execution unit performs molecular simulations for each of the initial conformations set by the initial conformation setting unit. The solubility feature amount calculation unit calculates a solubility feature amount based on the execution results of the molecular simulations by the molecular simulation execution unit. The partial three-dimensional structure information acquisition unit calculates partial three-dimensional structure information representing one or more partial structures of the molecule for each of the molecular simulations. The partial structure identification unit identifies the partial structure corresponding to the variation in the solubility feature amount based on the solubility feature amount calculated by the solubility feature amount calculation unit and the partial three-dimensional structure information calculated by the partial three-dimensional structure information acquisition unit. The thermodynamic stability evaluation unit classifies the solubility feature amount based on the partial structure identified by the partial structure identification unit to create groups, and evaluates the thermodynamic stability for each group. The variation evaluation unit selects one or several of the groups based on the evaluation by the thermodynamic stability evaluation unit, and calculates the variation in the solubility feature amount in the selected groups. The variation determination unit causes the initial conformation setting unit to set a new plurality of initial conformations when the variation calculated by the variation evaluation unit is equal to or greater than the target value of the variation. The output unit outputs an evaluation result of solubility based on the solubility feature amount used for the calculation of the variation when the variation calculated by the variation evaluation unit is less than the target value of the variation.

Advantages of the Invention

[0010] According to the present invention, it is possible to appropriately evaluate the molecular simulation results for the initial conformations and suppress the variation in the calculation results of the solubility feature amount due to differences in the initial conformations.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0012] Hereinafter, embodiments of a solubility evaluation apparatus, a solubility evaluation method, and a solubility evaluation program will be described with reference to the drawings. In addition, the same elements are denoted by the same reference numerals, and overlapping descriptions are omitted as appropriate. Also, the respective embodiments can be appropriately combined within a non-contradictory range.

[0013] [Overall Configuration] FIG. 1 is a block diagram of a solubility evaluation apparatus according to an embodiment. The solubility evaluation apparatus 1 is, for example, an apparatus that evaluates the solubility of a compound in a combination of a solute and a solvent used in flow synthesis.

[0014] For example, in flow synthesis, since a narrow flow path of several millimeters or less is used, blockage of the flow path due to the low solubility of reactants and products, which is a fundamental main factor, becomes a major operational problem. Generally, the solubility of a compound is determined by actually conducting a solubility test and quantifying the concentration when the solution becomes saturated.

[0015] On the one hand, when multiple reactions are required to fully synthesize a final product such as peptide synthesis, the number of combinations of solute and solvent increases exponentially with the number of reaction steps. In addition, since it is necessary to prepare a certain amount or more of the solute in advance for the solubility test, it is necessary to separately synthesize the intermediate for each reaction step. Therefore, it is not practical to conduct solubility tests on all possible intermediates that may occur in flow synthesis from the viewpoints of development man-hours and the time spent on the tests. Thus, a solubility prediction technique that does not use solubility tests is desired.

[0016] In addition, the solubility prediction method using computational chemistry is a method for analyzing the solvent effect due to the intermolecular interaction between the solute and the solvent based on statistical thermodynamics theory, and calculates solubility characteristic quantities including solvation free energy. Quantum science calculations such as molecular dynamics method, molecular orbital method, and density functional method are used as calculation methods in solubility prediction. In order to accurately calculate physical quantities including solubility characteristic quantities by calculation, it is desirable to reduce the variation in the calculation results of solubility characteristic quantities depending on the initial conformation. In particular, since solubility characteristic quantities including solvation free energy are physical quantities determined by the interaction of multiple molecular species of solute and solvent, the value may change significantly due to the steric difference of some partial structures. Therefore, the introduction of a method for specifying the partial structure corresponding to the variation in the calculation results of solubility characteristic quantities is expected to contribute to the evaluation and reduction of the variation derived from the initial conformation dependence of solubility characteristic quantities.

[0017] Therefore, the solubility evaluation apparatus 1 according to the present embodiment realizes a solubility prediction technique that does not use a solubility test and reduces the variation in the calculation results of solubility characteristic quantities derived from the initial conformation dependence of solubility characteristic quantities. Hereinafter, the variation in the calculation results of solubility characteristic quantities will be referred to as the variation in solubility characteristic quantities.

[0018] As shown in FIG. 1, the solubility evaluation apparatus 1 includes an initial conformation setting unit 11, a molecular simulation execution unit 12, a solubility characteristic quantity calculation unit 13, a partial three-dimensional structure information acquisition unit 14, a partial structure specification unit 15, a thermodynamic stability evaluation unit 16, a variation evaluation unit 17, a variation determination unit 18, and an output unit 19.

[0019] The initial conformation setting unit 11 has information on the solvent and solute that are the objects of solubility evaluation. For example, when synthesizing a compound in a plurality of reaction steps using flow synthesis, the initial conformation setting unit 11 holds information on the solvent and solute combinations that can be used for each reaction step. In practice, the initial conformation setting unit 11 performs the following initial conformation settings for all combinations of solvents and solutes. However, in the following description, a specific solvent and a specific solute, which are one of them, will be described.

[0020] The initial conformation setting unit 11 sets a solution system composed of solute molecules and solvent molecules together with the number of molecules, and sets a plurality of different initial conformations. The initial conformation setting unit 11 generates the initial conformation of each molecule by a random number. For example, the initial conformation setting unit 11 generates different initial conformations by varying the structure of the functional groups in the target molecule. More specifically, for example, in the case of a molecule having a carboxyl group as a functional group, the initial conformation setting unit 11 can generate different initial conformations by varying the relative position of the hydroxyl group with respect to the double bond between carbon and oxygen. Then, the initial conformation setting unit 11 outputs the generated initial conformation group to the molecular simulation execution unit 12.

[0021] Thereafter, when it is determined that the variation is large by the variation evaluation of the solubility characteristic quantity described below, the initial conformation setting unit 11 receives an instruction to re-set the initial conformation from the variation determination unit 18. Then, the initial conformation setting unit 11 randomly generates a plurality of initial conformations other than the previously set initial conformation group, and outputs them to the molecular simulation execution unit 12 in addition to the previously set initial conformation group. In this way, until the variation of the solubility characteristic quantity becomes small, the initial conformation setting unit 11 increases the initial conformation and repeats the molecular simulation.

[0022] The molecular simulation execution unit 12 receives the input of the information of the initial conformation group from the initial conformation setting unit 11. Next, the molecular simulation execution unit 12 performs a molecular simulation for each initial conformation. Then, the molecular simulation execution unit 12 outputs the results of the molecular simulation for each initial conformation to the solubility feature amount calculation unit 13. In addition, the molecular simulation execution unit 12 outputs the information of the initial conformation used in each molecular simulation to the partial three-dimensional structure information acquisition unit 14.

[0023] The solubility feature amount calculation unit 13 receives the input of the results of the molecular simulation for each initial conformation from the molecular simulation execution unit 12. Then, the solubility feature amount calculation unit 13 calculates the solubility feature amount using the results of the molecular simulation. The solubility feature amount is an index indicating how well the solute and the solvent dissolve together.

[0024] The solubility feature amount calculation unit 13 can use, for example, the solvation free energy or the dissolution free energy as the solubility feature amount. For example, the solubility feature amount calculation unit 13 performs a molecular dynamics simulation as the molecular simulation, and uses the results of the molecular dynamics calculation to calculate free energies such as the solvation free energy and the dissolution free energy by the free energy calculation method. The solubility feature amount calculation unit 13 can use, for example, the thermodynamic integration method, the free energy perturbation method, or the energy representation method as the free energy calculation method. Then, the solubility feature amount calculation unit 13 outputs the solubility feature amount, which is the calculation result for each molecular simulation, to the partial structure identification unit 15.

[0025] The partial three-dimensional structure information acquisition unit 14 receives, from the molecular simulation execution unit 12, the input of the information on the initial conformation used in the simulation for each molecular simulation. Then, the partial three-dimensional structure information acquisition unit 14 calculates the partial three-dimensional structure information representing the three-dimensional partial structure of the molecule in the molecular simulation result for each initial conformation. The partial three-dimensional structure information may include the information of one partial structure or the information of a plurality of partial structures. In this way, the partial three-dimensional structure information acquisition unit 14 acquires, for each initial conformation, a partial three-dimensional structure of a part rather than the whole of the molecule.

[0026] For example, the partial three-dimensional structure information acquisition unit 14 can use, as the partial three-dimensional structure information, information on angles in three-dimensional space such as the dihedral angle distribution, the trihedral angle distribution, and the bond angle of the partial structure of the molecule. In addition, the partial three-dimensional structure information acquisition unit 14 may use, as the partial three-dimensional structure information, the distance between atoms within or between the partial structures of the molecule. The partial three-dimensional structure information acquisition unit 14 can calculate a plurality of pieces of partial three-dimensional structure information according to the number of molecular structures of the target molecule.

[0027] FIG. 2 is a diagram showing an example of the partial three-dimensional structure information. For example, the partial three-dimensional structure information acquisition unit 14 acquires, as the partial three-dimensional structure information, the dihedral angle distributions of N partial structures from partial structure P1 to PN shown in FIG. 2 for the result of one molecular simulation. The graphs of partial structures P1 to PN in FIG. 2 represent the dihedral angle on the horizontal axis and the probability distribution on the vertical axis. Partial structures P1 to PN each represent a different partial structure of one molecule. Hereinafter, the case where the partial three-dimensional structure information acquisition unit 14 calculates the dihedral angle distribution as the partial three-dimensional structure information of the initial conformation of the target molecule will be described.

[0028] In this way, the partial three-dimensional structure information acquisition unit 14 calculates one or a plurality of dihedral angle distributions corresponding to each of the results of the molecular simulation. Then, the partial three-dimensional structure information acquisition unit 14 outputs the dihedral angle distribution, which is the partial three-dimensional structure information of the initial conformation for each result of the molecular simulation, to the partial structure identification unit 15.

[0029] The partial structure specifying unit 15 receives, from the solubility feature amount calculation unit 13, the input of the solubility feature amount for each result of the molecular simulation. Further, the partial structure specifying unit 15 receives, from the partial three-dimensional structure information acquisition unit 14, the input of the dihedral angle distribution which is the partial three-dimensional structure information of the initial conformation for each result of the molecular simulation. Then, the partial structure specifying unit 15 specifies, by performing multiple regression analysis with the partial three-dimensional structure information as the explanatory variable and the solubility feature amount as the objective variable, the partial structure corresponding to the variation in the solubility feature amount that occurs depending on the initial conformation. For example, the partial structure specifying unit 15 performs multiple regression analysis by statistical analysis or machine learning using the average value of the dihedral angle distribution as the explanatory variable, and specifies the partial structure based on the dihedral angle distribution correlated with the solubility feature amount.

[0030] FIG. 3 is a diagram showing an example of the relationship between the initial conformation, the solubility feature amount, and the partial three-dimensional structure information. In FIG. 3, the solubility feature amount is represented as Δμ. As shown in Table 101 of FIG. 3, for each initial conformation, there exist the solubility feature amount Δμ and the partial structures P1 to PN. Below, the specification of the partial structure using the dihedral angle distribution by the partial structure specifying unit 15 will be specifically described.

[0031] The partial structure specifying unit 15 calculates the average value of the dihedral angle distribution for each of the partial structures P1 to PN. Then, the partial structure specifying unit 15 specifies the partial structure corresponding to the variation in the solubility feature amount Δμ based on the calculated average value of the dihedral angle distribution. For example, the partial structure specifying unit 15 sets the objective variable as the solubility feature amount Δμ and the explanatory variable as the average value of the dihedral angle distribution for each partial structure, and selects the partial structure with the largest regression coefficient in the multiple regression analysis. Also, the partial structure specifying unit 15 may select a plurality of partial structures corresponding to the variation in the solubility feature amount Δμ. For example, it may select a predetermined number of partial structures having a regression coefficient equal to or greater than a predetermined regression coefficient threshold. A large regression coefficient indicates a large influence on the objective variable, and by selecting the partial structure with a large regression coefficient, the partial structure specifying unit 15 can select the partial structure that has a large influence on the variation in the solubility feature amount Δμ.

[0032] Next, as shown in the classification items of Table 101, the partial structure specifying unit 15 classifies the results of molecular simulation for each initial conformation based on the average value of the dihedral angle of the partial structure corresponding to the value of the solubility feature amount Δμ. For example, the partial structure specifying unit 15 selects the partial structure P3 as the partial structure corresponding to the variation of the solubility feature amount Δμ for each initial conformation in Table 101. In this case, the partial structure specifying unit 15 specifies that the value of the partial structure P3 of the initial conformations a and c is 0, and the value of the partial structure P3 of the initial conformations b and d is 180. Then, the partial structure specifying unit 15 classifies the results of the molecular simulation of the initial conformations a and c into group A, and classifies the results of the molecular simulation of the initial conformations b and d into group B. Thereafter, the partial structure specifying unit 15 outputs the classification result of the results of the molecular simulation of the initial conformations to the thermodynamic stability evaluation unit 16. Further, the partial structure specifying unit 15 outputs the information of the initial conformations of the results of each molecular simulation to the thermodynamic stability evaluation unit 16.

[0033] The thermodynamic stability evaluation unit 16 receives the input of the classification result of the results of the molecular simulation and the information of the initial conformations of the results of each molecular simulation from the partial structure specifying unit 15. Next, the thermodynamic stability evaluation unit 16 evaluates the thermodynamic stability for each classified group. For example, the thermodynamic stability evaluation unit 16 calculates the thermodynamic stability by the sum of the solvation free energy and the internal free energy of the molecule.

[0034] This thermodynamic stability is a value related to the ease of formation of substructures for each classification. The solvation free energy is a value based on the solubility feature quantity Δμ. The solvation free energy can represent the stability of the interaction between the solute and the solvent. Also, the internal free energy of the molecule is a value obtained by performing a molecular simulation alone on the solute molecule and represents the stability of the three-dimensional structure of the solute. In this case, the greater the thermodynamic stability in the negative direction, the higher the stability, indicating that the molecules in that state are more likely to be realized. That is, the thermodynamic stability evaluation unit 16 can select the molecular simulation result with a partial three-dimensional structure of a molecule that is more likely to be realized by selecting the partial three-dimensional structure of a molecule that is determined to have higher thermodynamic stability. The thermodynamic stability evaluation unit 16 outputs the calculated information on the thermodynamic stability for each classified group to the variation evaluation unit 17.

[0035] The variation evaluation unit 17 receives the input of the classification result of the molecular simulation result and the information on the thermodynamic stability for each group from the thermodynamic stability evaluation unit 16. Next, the variation evaluation unit 17 selects the group with the most excellent thermodynamic stability evaluated by the thermodynamic stability evaluation unit 16 as the group having the most stable structure. By selecting the group with the most excellent thermodynamic stability, the variation evaluation unit 17 can select the group with the highest realizability, that is, the group having the partial three-dimensional structure of the molecule that is most likely to exist when the solvent and the solute are mixed.

[0036] Then, the variation evaluation unit 17 evaluates the variation σ Δμ of the solubility feature quantity Δμ in the group having the most stable structure. For example, the variation evaluation unit 17 can use, as σ Δμ , the standard error derived from the difference in the initial conformations in the group having the most stable structure. After that, the variation evaluation unit 17 outputs the variation σ Δμ of the solubility feature quantity Δμ of the group having the most stable structure and the result of the molecular simulation to the variation determination unit 18.

[0037] Here, in the present embodiment, the variation evaluation unit 17 selects a group having the most stable structure and evaluates the variation σ of the solubility feature amount Δμ Δμ However, several groups may be selected. For example, if the thermodynamic stability of the variation evaluation unit 17 is superior to a predetermined stability threshold value, it is determined that the thermodynamic stability is sufficiently stable, and a group having a thermodynamic stability superior to the stability threshold value may be selected. Alternatively, the variation evaluation unit 17 may select a predetermined number of groups such as the top two or three groups.

[0038] FIG. 4 is a diagram showing the evaluation result of the variation of the solubility feature amount. In Table 102 of FIG. 4, the thermodynamic stability is represented by E. As shown in Table 102, when the thermodynamic stability E of each group is obtained, the variation evaluation unit 17 selects the group B having the largest thermodynamic stability E in the negative direction as the group having the most stable structure. Then, the variation evaluation unit 17 calculates the variation σ of the solubility feature amount Δμ in group B as 1.7. Here, in FIG. 4, the variations σ of the solubility feature amounts Δμ of other groups are also shown. However, the variation evaluation unit 17 only needs to calculate the variation σ of the solubility feature amount Δμ of the group having the most stable structure, and does not need to calculate the variations σ of the solubility feature amounts Δμ of other groups. Δμ Here, in FIG. 4, the variations σ of the solubility feature amounts Δμ of other groups are also shown. However, the variation evaluation unit 17 only needs to calculate the variation σ of the solubility feature amount Δμ of the group having the most stable structure, and does not need to calculate the variations σ of the solubility feature amounts Δμ of other groups. Δμ Also shown, but the variation evaluation unit 17 only needs to calculate the variation σ of the solubility feature amount Δμ of the group having the most stable structure, and does not need to calculate the variations σ of the solubility feature amounts Δμ of other groups. Δμ For the group having the most stable structure, the variation σ of the solubility feature amount Δμ is calculated, and the variation σ of the solubility feature amount Δμ of other groups Δμ Need not be calculated.

[0039] The variation determination unit 18 receives the variation σ of the solubility feature amount Δμ of the group having the most stable structure and the input of the result of the molecular simulation from the variation evaluation unit 17. The variation determination unit 18 has a variation target value σ Δμ Δμ-set in advance. Then, the variation determination unit 18 determines whether the variation σ of the solubility feature amount Δμ of the group having the most stable structure obtained by the evaluation of the variation evaluation unit 17 Δμ is less than the variation target value σ Δμ-set Δμ

[0040] The variation σ of the solubility feature amount Δμ of the group having the most stable structure Δμ ​is the target value σ of variation Δμ-set If it is above, the variation determination unit 18 determines that the variation of the solubility characteristic amount Δμ is large. Then, the variation determination unit 18 instructs the initial conformation setting unit 11 to reset the initial conformation.

[0041] On the other hand, the variation σ of the solubility characteristic amount Δμ of the group having the most stable structure Δμ is the target value σ of variation Δμ-set If it is above, the variation determination unit 18 determines that the variation of the solubility characteristic amount Δμ is small. Then, the variation determination unit 18 outputs the information of the solubility characteristic amount Δμ to the output unit 19 together with the information of the solvent and the solute.

[0042] Here, in the present embodiment, the initial conformations are increased randomly, but other methods may be used to increase the initial conformations. For example, the variation determination unit 18 may instruct the initial conformation setting unit 11 to reset the initial conformation together with the information of the group for which the information of the initial conformation is to be increased as the target group for increasing the information of the initial conformation of the group having the most stable structure. Thereby, the initial conformation setting unit 11 can selectively increase the initial conformations having a specific structure such as the structure of the group having the most stable structure.

[0043] The output unit 19 receives the input of the solubility characteristic amount of the group having the most stable structure from the variation determination unit 18 together with the information of the solvent and the solute. Then, the output unit 19 provides it to the user, for example, by causing the display device to display the evaluation result of the solubility based on the solubility characteristic amount of the group having the most stable structure corresponding to the information of the solvent and the solute.

[0044] The output unit 19 can provide information representing the relationship between the solvent and the solute, such as the solvation free energy, as the solubility characteristic amount. For example, the output unit 19 may cause the display device to display the solvation free energy for each solvent with respect to a certain solute in a comparable manner.

[0045] [Flow of Solubility Evaluation Process] FIG. 5 is a flowchart of the failure response work support process by the solubility evaluation apparatus according to the embodiment. Next, with reference to FIG. 5, the flow of the solubility evaluation process by the solubility evaluation apparatus 1 according to the present embodiment will be described.

[0046] The initial conformation setting unit 11 sets a solution system composed of solute molecules and solvent molecules together with the number of molecules using random numbers, and sets a plurality of different initial conformations (step S1).

[0047] The molecular simulation execution unit 12 acquires the initial conformation group generated by the initial conformation setting unit 11. Next, the molecular simulation execution unit 12 executes the calculation of the molecular simulation for each initial conformation (step S2).

[0048] The solubility feature amount calculation unit 13 receives the input of the result of the molecular simulation for each initial conformation from the molecular simulation execution unit 12. Then, the solubility feature amount calculation unit 13 calculates the solubility feature amount Δμ using the result of the molecular simulation (step S3).

[0049] The partial three-dimensional structure information acquisition unit 14 receives the input of the information of the initial conformation used for each molecular simulation from the molecular simulation execution unit 12. Then, the partial three-dimensional structure information acquisition unit 14 calculates partial three-dimensional structure information representing a part of the structure of the molecule in the molecular simulation result for each result of the molecular simulation (step S4).

[0050] The partial structure identification unit 15 receives the input of the solubility feature amount for each result of the molecular simulation from the solubility feature amount calculation unit 13. The partial structure identification unit 15 also receives the input of the partial three-dimensional structure information in the result of the molecular simulation for each initial conformation from the partial three-dimensional structure information acquisition unit 14. Then, the partial structure identification unit 15 identifies the partial structure correlated with the variation of the solubility feature amount Δμ from the partial three-dimensional structure information. Next, the partial structure identification unit 15 classifies the results of the molecular simulation for each initial conformation based on the partial structure corresponding to the value of the solubility feature amount Δμ (step S5).

[0051] The thermodynamic stability evaluation unit 16 receives, from the partial structure identification unit 15, the classification result of the molecular simulation results and information on the initial conformations of the results of each molecular simulation. Next, the thermodynamic stability evaluation unit 16 evaluates the thermodynamic stability for each classified group (step S6).

[0052] The variation evaluation unit 17 receives, from the thermodynamic stability evaluation unit 16, the classification result of the molecular simulation results and information on the thermodynamic stability for each group. Next, the variation evaluation unit 17 selects, as the group having the most stable structure, the group with the most excellent thermodynamic stability evaluated by the thermodynamic stability evaluation unit 16. Then, the variation evaluation unit 17 calculates the variation σ Δμ of the solubility characteristic quantity Δμ in the group having the most stable structure (step S7).

[0053] The variation determination unit 18 receives, from the variation evaluation unit 17, the variation σ Δμ of the group having the most stable structure and the input of the molecular simulation results. Next, the variation determination unit 18 determines whether the variation σ Δμ of the group having the most stable structure is less than the target value σ Δμ-set of the variation (step S8).

[0054] When the variation σ Δμ of the solubility characteristic quantity Δμ of the thermodynamic stability of the group having the most stable structure is greater than or equal to the target value σ Δμ-set of the variation (step S8: negative), the variation determination unit 18 instructs the initial conformation setting unit 11 to reset the initial conformation. The initial conformation setting unit 11, upon receiving the instruction to reset the initial conformation from the variation determination unit 18, randomly generates a plurality of new initial conformations, adds them to the previously generated initial conformation group, and newly sets the initial conformation (step S9). Thereafter, the solubility evaluation process returns to step S2.

[0055] On the contrary, when the variation σ Δμ of the solubility characteristic quantity Δμ of the group having the most stable structure is less than the target value σ Δμ-setIn the above case (step S8: affirmative), the variation determination unit 18 determines that the variation in the solubility characteristic quantity Δμ is small. Then, the variation determination unit 18 outputs the result of the molecular simulation to the output unit 19 together with the information on the solvent and the solute. The output unit 19 receives the input of the solubility characteristic quantity Δμ of the group having the most stable structure from the variation determination unit 18 together with the information on the solvent and the solute. Then, the output unit 19 provides the user by, for example, causing a display device to display an evaluation result of solubility based on the solubility characteristic quantity Δμ of the group having the most stable structure corresponding to the information on the solvent and the solute (step S10).

[0056] [Effect] As described above, the solubility evaluation apparatus according to the embodiment executes a molecular simulation for evaluating solubility by setting a plurality of initial conformations for a combination of a solvent and a solute, and classifies the results of the molecular simulation by partial structures of a part of the molecule. Next, the solubility evaluation apparatus according to the embodiment identifies the group having the highest thermodynamic stability among the classified groups, and evaluates the variation in the results of the molecular simulation in that group. Then, if the variation is equal to or greater than the threshold value, after re-setting by adding an initial conformation or the like, the molecular simulation and the evaluation of the variation in the results of the molecular simulation are repeated until the variation falls below the threshold value.

[0057] Thereby, it is possible to execute a molecular simulation using an initial conformation that is highly likely to exist and suppresses the variation in the results of the molecular simulation, and obtain the results of the molecular simulation. Thereby, it is possible to appropriately evaluate the molecular simulation results with respect to the initial conformation, suppress the variation in the calculation results due to the difference in the initial conformation, and perform an appropriate solubility evaluation. Therefore, the accuracy of solubility prediction using computational chemistry can be improved.

[0058] FIG. 6 is a diagram showing an example of the evaluation result of the solubility characteristic quantity provided by the solubility evaluation apparatus according to the embodiment. Graph 111 in FIG. 6 is a diagram showing the evaluation result of the solubility characteristic quantity when the initial conformation is not selected. Further, graph 112 is a diagram showing the evaluation result of the solubility characteristic quantity when the solubility evaluation apparatus 1 according to the present embodiment is used.

[0059] Both graphs 111 and 112 are graphs representing the solvation free energy of a solution composed of a specific solute and solvents #1 to #6 calculated by computational chemistry. In both graphs 111 and 112, the horizontal axis represents the type of solvent, and the vertical axis represents the solvation free energy. Also, both graphs 111 and 112 show that the stability improves as it goes downward. For example, the output unit 19 of the solubility evaluation apparatus 1 according to the present embodiment causes graph 112 to be displayed on the display device.

[0060] The line segments attached to the tips of the bands representing the solvation free energy of each of the solvents #1 to #6, such as line segment 121 in graph 111, represent the width of the variation in the calculated solvation free energy in each of the solvents #1 to #6. Similarly, each line segment such as line segment 122 in graph 112 also represents the width of the variation in the calculated solvation free energy in each of the solvents #1 to #6. The variation in graph 112 is less than that in graph 111. For example, in graph 111, there is a possibility that the solvation free energy of solvent #2 is more negative than that of solvent #4. In contrast, in graph 112, it can be confirmed that the solvation free energy of solvent #2 is less negative than that of solvent #4. Therefore, for example, when choosing either solvent #2 or solvent #4, by using graph 112 provided by the solubility evaluation apparatus 1 according to the present embodiment, it is possible to surely select a solvent with high stability. For example, when performing synthesis by 10 reaction steps, the stability in each reaction step can be appropriately evaluated, and it is possible to select an optimal route with high solubility.

[0061] Here, in FIG. 6, the solute was fixed to represent the dependence of solubility on the solvent. However, the solubility evaluation apparatus 1 according to the present embodiment can also provide information representing the dependence of solubility on the solute by fixing the solvent. However, when the solvent is fixed, it is conceivable that the solubility may change depending on the solid state of the solute. Therefore, when providing information representing the dependence of solubility on the solute by fixing the solvent, it is preferable that the solubility evaluation apparatus 1 evaluates the solubility characteristic amount in consideration of the solid state of the solute and provides an evaluation of the evaluation result.

[0062] In addition, as a method for analyzing the calculation results of molecular dynamics, a technique for classifying the calculation results as preferable or not using the fluctuation information (RMSD) of the three-dimensional structure of the molecule as an index can be considered. However, in this technique, RMSD representing the similarity of the overall three-dimensional structure of the molecule is used as the information on the three-dimensional structure of the molecule. Since RMSD is defined as the square root of the sum of the squares of the distances between corresponding two points in two three-dimensional structures, it is often used as an index of the similarity of the overall three-dimensional structure, but it is not suitable for the purpose of extracting differences in specific partial structures. Since the partial structure of the molecule greatly affects the variation in the solubility characteristic amount, in order to reduce the variation in the calculation results due to differences in the initial conformations, it is effective to identify the partial structure corresponding to the variation in the calculation results and classify each partial structure. If only classification using RMSD indicating the similarity of the whole molecule as the three-dimensional structure information is performed without considering the partial structure, it is difficult to classify according to the structure of the molecule according to the variation in the solubility characteristic amount.

[0063] On the other hand, the solubility evaluation apparatus according to the present embodiment identifies the three-dimensional partial structure of the molecule corresponding to the variation in the calculation results and classifies each partial structure, and it is easy to identify the structure for suppressing the variation in the solubility characteristic amount. Therefore, unlike the technique for classifying molecules using RMSD as an index, the solubility evaluation apparatus according to the present embodiment can appropriately evaluate the initial conformation, suppress the variation in the calculation results due to differences in the initial conformation, and perform an appropriate solubility evaluation.

[0064] [System] Regarding the processing procedures, control procedures, specific names, and information including various data and parameters shown in the above-mentioned documents and drawings, they can be arbitrarily changed unless otherwise specified.

[0065] In addition, each component of each illustrated device is a functional concept and does not necessarily need to be physically configured as shown in the figure. That is, the specific forms of dispersion and integration of each device are not limited to those shown in the figure. In other words, all or part of it can be functionally or physically dispersed and integrated in any unit according to various loads, usage situations, etc.

[0066] Furthermore, each processing function performed by each device can be realized in whole or in any part by a CPU (Central Processing Unit) and a program analyzed and executed by the CPU, or can be realized as hardware by wired logic.

[0067] [Hardware] Next, a hardware configuration example of the solubility evaluation device 1 will be described. FIG. 7 is a hardware configuration diagram of the solubility evaluation device. As shown in FIG. 7, the solubility evaluation device 1 includes a processor 91, a memory 92, a communication device 93, and an HDD (Hard Disk Drive) 94. The processor 91 is connected to the memory 92, the communication device 93, and the HDD 94 via a bus.

[0068] The communication device 93 is a network interface card or the like and is used for communication with other information processing devices.

[0069] The HDD 94 is an auxiliary storage device. The HDD 94 stores various programs including programs for realizing the functions of the initial conformation setting unit 11, the molecular simulation execution unit 12, the solubility feature amount calculation unit 13, the partial three-dimensional structure information acquisition unit 14, the partial structure identification unit 15, the thermodynamic stability evaluation unit 16, the variation evaluation unit 17, the variation determination unit 18, and the output unit 19.

[0070] The processor 91 reads out various programs stored in the HDD 94, expands them in the memory 92, and executes them. As a result, the processor 91 realizes the functions of the initial conformation setting unit 11, the molecular simulation execution unit 12, the solubility feature amount calculation unit 13, the partial three-dimensional structure information acquisition unit 14, the partial structure identification unit 15, the thermodynamic stability evaluation unit 16, the variation evaluation unit 17, the variation determination unit 18, and the output unit 19.

[0071] As described above, the solubility evaluation apparatus 1 operates as an information processing apparatus that executes various processing methods by reading and executing a program. Further, the solubility evaluation apparatus 1 can also realize the same functions as those of the above-described embodiments by reading the program from the recording medium by the medium reading device and executing the read program. Note that the program described here is not limited to being executed by the solubility evaluation apparatus 1. For example, the present invention can be similarly applied when another computer or server executes the program, or when these cooperate to execute the program.

[0072] This program can be distributed via a network such as the Internet. Further, this program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, a MO (Magneto-Optical disk), a DVD (Digital Versatile Disc), and can be executed by being read from the recording medium by a computer.

[0073] How to describe some examples of combinations of the disclosed technical features.

[0074] (1) An initial conformation setting unit that sets a plurality of different initial conformations of a molecule, A molecular simulation execution unit that executes a molecular simulation for each of the initial conformations set by the initial conformation setting unit, A solubility feature amount calculation unit that calculates a solubility feature amount based on the execution result of the molecular simulation by the molecular simulation execution unit; A partial three-dimensional structure information acquisition unit that calculates partial three-dimensional structure information representing one or more three-dimensional partial structures of the molecule for each of the molecular simulations; A partial structure identification unit that identifies the partial structure corresponding to the variation in the solubility feature amount based on the solubility feature amount calculated by the solubility feature amount calculation unit and the partial three-dimensional structure information calculated by the partial three-dimensional structure information acquisition unit; A thermodynamic stability evaluation unit that classifies the solubility feature amount based on the partial structure identified by the partial structure identification unit to create groups, and evaluates the thermodynamic stability for each group; A variation evaluation unit that selects one or several of the groups based on the evaluation by the thermodynamic stability evaluation unit, and calculates the variation in the solubility feature amount in the selected group; A variation determination unit that causes the initial conformation setting unit to set a plurality of new initial conformations when the variation calculated by the variation evaluation unit is equal to or greater than the target value of the variation; An output unit that outputs an evaluation result of solubility based on the solubility feature amount used for calculating the variation when the variation calculated by the variation evaluation unit is less than the target value of the variation. A solubility evaluation apparatus, characterized by comprising the above. (2) The molecular simulation execution unit uses molecular dynamics simulation, and the solubility evaluation apparatus according to (1). (3) The solubility feature amount calculation unit uses solvation free energy as the solubility feature amount, and the solubility evaluation apparatus according to (1) or (2). (4) The partial three-dimensional structure information acquisition unit uses a three-dimensional structure related to an angle as the partial three-dimensional structure information, and the solubility evaluation apparatus according to any one of (1) to (3). (5) The partial three-dimensional structure information acquisition unit uses a dihedral angle distribution as the partial three-dimensional structure information, and the solubility evaluation apparatus according to (4). (6) The partial structure specifying unit specifies the partial structure by performing a regression analysis using the partial three-dimensional structure information as an explanatory variable and the solubility feature amount as an objective variable, and the solubility evaluation apparatus according to any one of (1) to (5). (7) The partial structure specifying unit performs the regression analysis by statistical analysis or machine learning using the average value of the dihedral angle distribution as the partial three-dimensional structure information as an explanatory variable, and specifies the partial structure based on the dihedral angle distribution correlated with the solubility feature amount, and the solubility evaluation apparatus according to (6). (8) The variation evaluation unit selects the group evaluated as the most thermodynamically stable by the thermodynamic stability evaluation unit, and the solubility evaluation apparatus according to any one of (1) to (8). (9) The initial conformation setting unit adds additional initial conformations belonging to the group selected by the variation evaluation unit to the plurality of initial conformations to set the new plurality of initial conformations, and the solubility evaluation apparatus according to any one of (1) to (8). (10) A solubility evaluation apparatus sets a plurality of different initial conformations of a molecule, performs a molecular simulation for each of the set initial conformations, calculates a solubility feature amount based on the execution result of the molecular simulation, calculates partial three-dimensional structure information representing one or a plurality of three-dimensional partial structures of the molecule for each of the molecular simulations, based on the solubility feature amount and the partial three-dimensional structure information, specifies the partial structure corresponding to the variation of the solubility feature amount, creates a group by classifying the solubility feature amount based on the specified partial structure, evaluates the thermodynamic stability for each group, Based on the evaluation, select one or several of the groups, calculate the variation of the solubility characteristic quantity in the selected group, When the calculated variation is equal to or greater than the target value of the variation, set a plurality of new initial conformations, and execute the molecular simulation, calculate the solubility characteristic quantity, calculate the partial three-dimensional structure information, identify the partial structure, evaluate the thermodynamic stability, and calculate the variation based on the plurality of new initial conformations, When the calculated variation is less than the target value of the variation, output the solubility evaluation result based on the solubility characteristic quantity used for the calculation of the variation in that case A solubility evaluation method characterized by executing the process. (11) Set a plurality of different initial conformations of the molecule, Execute a molecular simulation for each of the set initial conformations, Based on the execution result of the molecular simulation, calculate the solubility characteristic quantity, Calculate the partial three-dimensional structure information representing one or more three-dimensional partial structures of the molecule for each of the molecular simulations, Based on the solubility characteristic quantity and the partial three-dimensional structure information, identify the partial structure corresponding to the variation of the solubility characteristic quantity, Classify the solubility characteristic quantity based on the identified partial structure to create groups, evaluate the thermodynamic stability for each group, Based on the evaluation, select one or several of the groups, calculate the variation of the solubility characteristic quantity in the selected group, When the calculated variation is equal to or greater than the target value of the variation, set a plurality of new initial conformations, and execute the molecular simulation, calculate the solubility characteristic quantity, calculate the partial three-dimensional structure information, identify the partial structure, evaluate the thermodynamic stability, and evaluate the variation based on the plurality of new initial conformations, When the calculated variation is less than the target value of the variation, output the solubility evaluation result based on the solubility characteristic quantity used for the calculation of the variation in that case A work support program characterized by causing a computer to execute a process.

Explanation of symbols

[0075] 1 Solubility evaluation apparatus 11 Initial conformation setting unit 12 Molecular simulation execution unit 13 Solubility feature quantity calculation unit 14 Partial three-dimensional structure information acquisition unit 15 Partial structure identification unit 16 Thermodynamic stability evaluation unit 17 Variation evaluation unit 18 Variation determination unit 19 Output unit

Claims

1. An initial conformation setting unit that sets a plurality of different initial conformations of a molecule, A molecular simulation execution unit that performs molecular simulations for each of the initial conformations set by the initial conformation setting unit, A solubility feature amount calculation unit that calculates a solubility feature amount based on the execution result of the molecular simulation by the molecular simulation execution unit, A partial three-dimensional structure information acquisition unit that calculates partial three-dimensional structure information representing one or a plurality of three-dimensional partial structures of the molecule for each of the molecular simulations, A partial structure identification unit that identifies the partial structure corresponding to the variation in the solubility feature amount based on the solubility feature amount calculated by the solubility feature amount calculation unit and the partial three-dimensional structure information calculated by the partial three-dimensional structure information acquisition unit, A thermodynamic stability evaluation unit that classifies the solubility feature amount based on the partial structure identified by the partial structure identification unit to create groups and evaluates the thermodynamic stability for each group, A variation evaluation unit that selects one or several of the groups based on the evaluation by the thermodynamic stability evaluation unit and calculates the variation in the solubility feature amount in the selected group, A variation determination unit that causes the initial conformation setting unit to set a new plurality of initial conformations when the variation calculated by the variation evaluation unit is equal to or greater than the target value of the variation, An output unit that outputs an evaluation result of solubility based on the solubility feature amount used for the calculation of the variation when the variation calculated by the variation evaluation unit is less than the target value of the variation A solubility evaluation apparatus characterized by comprising the above.

2. The molecular simulation execution unit uses molecular dynamics simulation. The solubility evaluation apparatus according to Claim 1, characterized in that.

3. The solubility feature amount calculation unit uses solvation free energy as the solubility feature amount. The solubility evaluation apparatus according to Claim 1, characterized in that.

4. The partial three-dimensional structure information acquisition unit uses a three-dimensional structure related to an angle as the partial three-dimensional structure information. The solubility evaluation apparatus according to Claim 1, characterized in that.

5. The partial three-dimensional structure information acquisition unit uses a dihedral angle distribution as the partial three-dimensional structure information. The solubility evaluation apparatus according to Claim 4, characterized in that.

6. The partial structure specifying unit specifies the partial structure by performing a regression analysis using the partial three-dimensional structure information as an explanatory variable and the solubility characteristic quantity as an objective variable, according to the solubility evaluation apparatus of claim 1.

7. The partial structure specifying unit performs the regression analysis by statistical analysis or machine learning using the average value of the dihedral angle distribution as an explanatory variable for the partial three-dimensional structure information, and specifies the partial structure based on the dihedral angle distribution correlated with the solubility characteristic quantity, according to the solubility evaluation apparatus of claim 6.

8. The variation evaluating unit selects the group evaluated to be the most thermodynamically stable by the thermodynamic stability evaluating unit, according to the solubility evaluation apparatus of claim 1.

9. The initial conformation setting unit adds additional initial conformations belonging to the group selected by the variation evaluating unit to the plurality of initial conformations to set the new plurality of initial conformations, according to the solubility evaluation apparatus of claim 1.

10. A solubility evaluation apparatus, sets a plurality of different initial conformations of a molecule, performs a molecular simulation for each of the set initial conformations, calculates a solubility characteristic quantity based on the execution result of the molecular simulation, calculates partial three-dimensional structure information representing one or more three-dimensional partial structures of the molecule for each of the molecular simulations, identifies the partial structure corresponding to the variation of the solubility characteristic quantity based on the solubility characteristic quantity and the partial three-dimensional structure information, classifies the solubility characteristic quantity based on the identified partial structure to create groups, evaluates the thermodynamic stability for each group, selects one or several of the groups based on the evaluation, calculates the variation of the solubility characteristic quantity in the selected group, when the calculated variation is equal to or greater than a target value of the variation, sets a new plurality of initial conformations and executes the processes of the molecular simulation, the calculation of the solubility characteristic quantity, the calculation of the partial three-dimensional structure information, the identification of the partial structure, the evaluation of the thermodynamic stability, and the calculation of the variation based on the new plurality of initial conformations, when the calculated variation is less than the target value of the variation, outputs a solubility evaluation result based on the solubility characteristic quantity used for the calculation of the variation in that case characterized by executing the process.

11. sets a plurality of different initial conformations of a molecule, Execute molecular simulation for each of the set initial conformations, calculate solubility characteristic quantities based on the execution results of the molecular simulation, calculate partial three-dimensional structure information representing one or more three-dimensional partial structures of the molecule for each of the molecular simulations, specify the partial structure corresponding to the variation in the solubility characteristic quantity based on the solubility characteristic quantity and the partial three-dimensional structure information, classify the solubility characteristic quantity based on the specified partial structure to create groups, and evaluate the thermodynamic stability for each group, select one or several of the groups based on the evaluation, and calculate the variation in the solubility characteristic quantity in the selected groups, when the calculated variation is equal to or greater than the target value of the variation, set a plurality of new initial conformations, and execute the processes of the molecular simulation, the calculation of the solubility characteristic quantity, the calculation of the partial three-dimensional structure information, the specification of the partial structure, the evaluation of the thermodynamic stability, and the evaluation of the variation based on the plurality of new initial conformations, when the calculated variation is less than the target value of the variation, output a solubility evaluation result based on the solubility characteristic quantity used for the calculation of the variation in that case A solubility evaluation program characterized by causing a computer to execute the process.

Images