JP2024139949A - EVALUATION PROGRAM, EVALUATION APPARATUS, AND EVALUATION METHOD - Google Patents

Abstract

To efficiently evaluate a structure of a compound.SOLUTION: An evaluation program causes a computer to execute processing including: dividing, a distribution of population variables calculated from a structure of a compound in predetermined environment into one or more clusters; determining a representative structure representing the cluster from the structures included in each of the clusters; and calculating a score based on ratio of the structures present in a range of predetermined distance from the representative structure among the structures included in the distribution.SELECTED DRAWING: Figure 3

Description

This disclosure relates to an evaluation program, an evaluation device, and an evaluation method.

In the field of drug discovery, a search is conducted for lead compounds with improved physical properties. The search for lead compounds is conducted by evaluating the physical properties of modified compounds that are made by modifying a part of a hit compound that binds to a target protein.

For example, Patent Document 1 discloses a complex structure prediction device that calculates a docking score that represents an interaction energy according to the relative arrangement of a first input structure and a second input structure to be docked.

JP 2008-287529 A

However, conventional techniques have the problem that evaluating the physical properties of compounds requires high time and financial costs. Since compounds with similar structures are expected to have similar physical properties, there is a demand for methods to evaluate the structural characteristics of compounds.

In one aspect, the aim is to efficiently evaluate the structure of compounds.

According to one embodiment, the evaluation program causes a computer to execute a process of dividing a distribution of collective variables calculated from the structure of a compound in a specified environment into one or more clusters, determining a representative structure representing each cluster from the structures contained in each cluster, and calculating a score based on the proportion of structures present within a specified distance from the representative structure among the structures contained in the distribution.

The structure of compounds can be evaluated efficiently.

1 is a block diagram showing an example of the overall configuration of a molecular structure evaluation system. FIG. 2 is a block diagram showing an example of a hardware configuration of a computer. FIG. 2 is a block diagram showing an example of a functional configuration of an evaluation device. FIG. 13 is a diagram illustrating an example of a score function. FIG. 13 is a diagram illustrating a first example of the mean square deviation. FIG. 13 is a diagram illustrating a second example of the mean square deviation. FIG. 13 is a diagram showing a first example of a stability score. FIG. 13 is a diagram showing a second example of a stability score. 1 is a flowchart showing an example of an evaluation method. FIG. 1 shows an example of a prediction of an increase or decrease in binding activity.

Hereinafter, each embodiment of the present invention will be described with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configuration are designated by the same reference numerals, and redundant description will be omitted.

[Embodiment]
Computer-assisted new drug development is called drug discovery. The field of drug discovery is undergoing a major transformation due to the boom in big data and dedicated computers. In drug discovery, a lead compound with improved physical properties is searched for by modifying a part (e.g., amino acid residues) of a known hit compound that binds to a target protein. In drug discovery, for example, it is required to improve binding activity, membrane permeability, heat resistance, etc.

Conventionally, the binding activity of modified compounds designed from hit compounds was determined through experiments. In these experiments, the modified compounds were synthesized and their binding activity was measured, resulting in high time and financial costs. Computer-assisted new drug development is also underway, but simulations with extremely high computational costs are required. For example, this requires obtaining the complex structure of the target molecule bound to the hit compound, simulating the interaction, simulating the change to a modified compound, and calculating the binding free energy to evaluate the binding activity.

It is extremely difficult to determine the complex structure through computational experiments, and X-ray experiments and other methods must be used in conjunction. In addition, since compounds that are potential drug candidates are often in a charged state in water, calculating the binding free energy is also extremely difficult.

In conventional drug discovery, the mainstream approach is to calculate the ranking of binding activity expressed by a simple score function in order to reduce calculation costs. However, conventional simple score functions are often insufficiently accurate and show little correlation with experimental values. For this reason, the contribution of conventional score functions to the efficiency of drug discovery is limited.

In this embodiment, a score function is introduced that quantifies the conformational characteristics of a compound. In one aspect, it is possible to evaluate the conformational characteristics of a compound in a solvent based on the score calculated by the score function of this embodiment.

By applying the score of this embodiment, a method for evaluating the similarity of the structure of modified compounds designed from hit compounds can be realized. In addition, by using the score of this embodiment, a method for evaluating whether a compound passes through a specific environment, such as membrane permeability, can be realized. Furthermore, by using the score of this embodiment, it is possible to predict the increase or decrease in binding activity, membrane permeability, heat resistance, etc. of modified compounds.

<Overall configuration of molecular structure evaluation system>
One embodiment of the present invention is a molecular structure evaluation system for evaluating the three-dimensional structure of a compound. The molecular structure evaluation system calculates the three-dimensional structure of the compound using a molecular simulation such as a molecular dynamics method, and calculates a score indicating the stability of the three-dimensional structure of the compound (hereinafter, also referred to as a "stability score").

The overall configuration of the molecular structure evaluation system in this embodiment will be described with reference to FIG. 1. FIG. 1 is a block diagram showing an example of the overall configuration of the molecular structure evaluation system in this embodiment.

As shown in FIG. 1, the molecular structure evaluation system 1 in this embodiment includes an evaluation device 10 and a user terminal 20. The evaluation device 10 and the user terminal 20 are connected to each other so as to be able to communicate data with each other via a communication network N1 such as a LAN (Local Area Network) or the Internet.

The evaluation device 10 is an information processing device such as a personal computer, a workstation, or a server that calculates a stability score for a compound. The evaluation device 10 receives structural information on a modified compound from a user terminal 20 and calculates a stability score for the modified compound. The evaluation device 10 transmits evaluation results including the stability score for the modified compound to the user terminal 20.

The structural information includes information indicating the three-dimensional structure of the modified compound to be evaluated, and information indicating the environment in which the physical properties of the modified compound are evaluated. The information indicating the environment is, for example, information indicating the three-dimensional structure of the solvent molecule. The solvent varies depending on the physical property to be evaluated. For example, when evaluating the binding activity or heat resistance of the modified compound, the solvent is water. Also, for example, when evaluating the membrane permeability of the modified compound, chloroform or other solvents can be used.

The user terminal 20 is an information processing terminal such as a personal computer, smartphone, or tablet terminal operated by a user of the molecular structure evaluation system 1. The user terminal 20 generates structural information regarding the modified compound in response to the user's operation and transmits it to the evaluation device 10. The user terminal 20 receives the evaluation results of the modified compound from the evaluation device 10 and outputs them to the user.

The overall configuration of the molecular structure evaluation system 1 shown in FIG. 1 is one example, and various system configuration examples are possible depending on the application and purpose. For example, one or more of the evaluation devices 10 and user terminals 20 may be included in the molecular structure evaluation system 1. For example, the evaluation device 10 may be realized by multiple computers, or may be realized as a cloud computing service. The classification of devices such as the evaluation device 10 and user terminal 20 shown in FIG. 1 is one example.

<Hardware configuration of molecular structure evaluation system>
The hardware configuration of each device included in the molecular structure evaluation system 1 in this embodiment will be described with reference to FIG.

<Computer hardware configuration>
The evaluation device 10 and the user terminal 20 in this embodiment are realized by, for example, a computer. Fig. 2 is a block diagram showing an example of the hardware configuration of the computer in this embodiment.

As shown in FIG. 2, the computer 500 in this embodiment has a CPU (Central Processing Unit) 501, a ROM (Read Only Memory) 502, a RAM (Random Access Memory) 503, a HDD (Hard Disk Drive) 504, an input device 505, a display device 506, a communication I/F (Interface) 507, and an external I/F 508. The CPU 501, the ROM 502, and the RAM 503 form a so-called computer. Each piece of hardware in the computer 500 is connected to each other via a bus line 509. Note that the input device 505 and the display device 506 may be connected to the external I/F 508 for use.

The CPU 501 is a computing device that reads programs and data from a storage device such as the ROM 502 or the HDD 504 onto the RAM 503 and executes processing to realize the overall control and functions of the computer 500. The computer 500 may have a GPU (Graphics Processing Unit) in addition to or instead of the CPU 501.

ROM 502 is an example of a non-volatile semiconductor memory (storage device) that can retain programs and data even when the power is turned off. ROM 502 functions as a main storage device that stores various programs, data, etc. required for CPU 501 to execute various programs installed in HDD 504. Specifically, ROM 502 stores boot programs such as BIOS (Basic Input/Output System) and EFI (Extensible Firmware Interface) that are executed when computer 500 is started, as well as data such as OS (Operating System) settings and network settings.

RAM 503 is an example of a volatile semiconductor memory (storage device) from which programs and data are erased when the power is turned off. RAM 503 is, for example, a dynamic random access memory (DRAM) or a static random access memory (SRAM). RAM 503 provides a working area into which various programs installed in HDD 504 are expanded when executed by CPU 501.

HDD 504 is an example of a non-volatile storage device that stores programs and data. The programs and data stored in HDD 504 include an OS, which is basic software that controls the entire computer 500, and applications that provide various functions on the OS. Note that instead of HDD 504, computer 500 may use a storage device that uses flash memory as a storage medium (e.g., SSD: Solid State Drive, etc.).

The input device 505 includes a touch panel that the user uses to input various signals, operation keys or buttons, a keyboard or mouse, a microphone for inputting sound data such as voice, etc.

The display device 506 is composed of a display such as a liquid crystal or organic EL (Electro-Luminescence) display for displaying a screen, a speaker for outputting sound data such as voice, etc.

The communication I/F 507 is an interface that connects to a communication network and enables the computer 500 to perform data communication.

The external I/F 508 is an interface with external devices. Examples of external devices include a drive device 510.

The drive device 510 is a device for setting the recording medium 511. The recording medium 511 here includes media that record information optically, electrically, or magnetically, such as CD-ROMs, flexible disks, and magneto-optical disks. The recording medium 511 may also include semiconductor memory that records information electrically, such as ROM and flash memory. This allows the computer 500 to read and/or write to the recording medium 511 via the external I/F 508.

The various programs to be installed in the HDD 504 are installed, for example, by setting the distributed recording medium 511 in a drive device 510 connected to the external I/F 508 and reading the various programs recorded on the recording medium 511 by the drive device 510. Alternatively, the various programs to be installed in the HDD 504 may be installed by downloading them from a network different from the communication network via the communication I/F 507.

<Functional configuration of molecular structure evaluation system>
The functional configuration of the molecular structure evaluation system in this embodiment will be described with reference to Fig. 3. Fig. 3 is a block diagram showing an example of the functional configuration of the molecular structure evaluation system 1 in this embodiment.

<Functional configuration of the evaluation device>
As shown in FIG. 3 , the evaluation device 10 in this embodiment includes an input unit 11, a structure analysis unit 12, a population variable calculation unit 13, a cluster division unit 14 (an example of a division unit), a representative structure determination unit 15 (an example of a determination unit), a score calculation unit 16 (an example of a calculation unit), and an output unit 17.

The input unit 11, structure analysis unit 12, population variable calculation unit 13, cluster division unit 14, representative structure determination unit 15, score calculation unit 16, and output unit 17 are realized, for example, by a process executed by the CPU 501 of a program loaded onto the RAM 503 from the HDD 504 shown in FIG. 2.

The input unit 11 accepts input of structural information regarding the modified compound to be evaluated. The input unit 11 accepts structural information by receiving the structural information from the user terminal 20. The input unit 11 may accept structural information by acquiring structural information input by the user to the input device 505 of the evaluation device 10.

The structure analysis unit 12 analyzes the three-dimensional structure of the modified compound based on the structural information input to the input unit 11. The structure analysis unit 12 first calculates the three-dimensional structure of the modified compound in a specified environment using a molecular simulation based on molecular dynamics. Next, the structure analysis unit 12 samples the three-dimensional structure of the modified compound generated by the molecular simulation at specified time intervals within a specified time range.

The time range for sampling the three-dimensional structure may be any length of time that allows sufficient samples to be obtained for evaluation. However, since the molecular structure is not stable immediately after starting the simulation, it is advisable to start sampling after a certain amount of time has passed. The time interval for sampling the three-dimensional structure may be determined arbitrarily.

In this embodiment, sampling begins 200 nanoseconds or more after the start of the simulation, and sampling is performed every 10 picoseconds for 200 nanoseconds. Therefore, in this embodiment, 20,000 samples of three-dimensional structures are sampled.

The collective variable calculation unit 13 calculates predetermined collective variables for each of the three-dimensional structures sampled by the structure analysis unit 12. The collective variable calculation unit 13 places the calculated collective variables in a structure space. The structure space is a space that includes all structures of the modified compound. In this way, the collective variable calculation unit 13 obtains the distribution of collective variables of the modified compound in a specified environment.

The collective variables in this embodiment are preferably collective variables capable of separating each three-dimensional structure in the structural space. It is preferable that the collective variables are capable of completely separating each three-dimensional structure in the structural space. In this embodiment, the root mean square deviation (RMSD) of the atomic positions of the main chain and Cβ is used as an example of the collective variables.

The cluster division unit 14 divides the distribution of the collective variables acquired by the collective variable calculation unit 13 into one or more clusters. In this embodiment, as an example, the distribution of the collective variables is clustered by a density peak clustering method. The number of clusters is not limited, but it is preferable that it is approximately 1 to 10. In this way, the cluster division unit 14 forms one or more clusters in the structural space.

The representative structure determination unit 15 determines a cluster center structure (an example of a representative structure) that represents each cluster formed by the cluster division unit 14. The representative structure determination unit 15 first divides the space contained in each cluster in the structural space into subspaces. The size of the subspace may be set arbitrarily. As an example, the size of the subspace in this embodiment is set to 1 angstrom.

The representative structure determination unit 15 also calculates the density of each subspace. In the subspace with the highest density, the representative structure determination unit 15 identifies the three-dimensional structure corresponding to the collective variable located closest to the center, and determines it as the cluster center structure.

The score calculation unit 16 calculates a stability score _Tcc based on the cluster center structure determined by the representative structure determination unit 15. The stability score _Tcc will be described in detail later.

The output unit 17 outputs the evaluation result including the stability score _Tcc calculated by the score calculation unit 16. When the output unit 17 receives structural information from the user terminal 20, it transmits the evaluation result to the user terminal 20. When the output unit 17 accepts structural information input to the input device 505, it displays the evaluation result on the display device 506.

<Details of Stability Score _Tcc >
In the field of drug discovery, after a hit compound (drug candidate compound) is found through experiments, etc., a part of the drug candidate compound is modified to investigate the site that interacts with the target molecule. Once the site that interacts with the target molecule (target protein) is identified, a compound with improved binding activity is searched for while maintaining the interaction and modifying other sites. In other words, the search for compounds is based on maintaining the binding site and binding pose of the drug candidate compound and the target molecule.

A hit compound has binding activity towards a target molecule. When a hit compound binds to a target molecule, the entropy of the hit compound in the solvent is lost. Therefore, in order to improve binding activity, it is important to keep the entropy as low as possible in the solvent. Since it is difficult to estimate the gain due to interaction with the target molecule, it is important to further stabilize the structure of the hit compound while maintaining the interacting site. Therefore, a score is required that can evaluate the similarity between the three-dimensional structure of a compound and that of the hit compound.

The stability score _Tcc is a score that quantifies the conformational characteristics of a compound. The stability score _Tcc is calculated by dividing the distribution of collective variables calculated from the conformation of the compound into clusters, and based on the proportion of structures that exist within a predetermined distance from a cluster center structure that represents each cluster.

Specifically, the stability score _Tcc in this embodiment is calculated by the score function _Tcc (r) of equation (1).

where i is the cluster number, r is a cluster variable, _rc is a predetermined cutoff radius, _{cc i} ^ref is the cluster center structure of cluster i, and _{ρ i} (r) is the proportion of cluster i.

The existence ratio ρ of cluster i is the ratio of collective variables r that exist within a range of a cutoff radius r _c from the cluster center structure cc _i ^ref to the entire collective variables r. The cutoff radius r _c can be set arbitrarily. In this embodiment, the cutoff radius r _c is, for example, 2 angstroms.

FIG. 4 is a diagram showing an example of a score function in this embodiment. As shown in FIG. 4, a plurality of clusters 310-1 to 310-3 are formed in the structural space 300. Cluster center structures cc ₁ ^ref to cc ₃ ^ref are determined for each of the clusters 310-1 to 310-3. Note that, for the sake of explanation, a two-dimensional structural space 300 is illustrated in FIG. 4, but the number of dimensions of the structural space is the same as the number of dimensions of the collective variables. For example, when the collective variable is the mean square deviation, the structural space is one-dimensional.

In cluster 310-1, a region 320-1 is defined within a cutoff radius r _c centered on the cluster center structure cc ₁ ^ref . The proportion of collective variables r included in the region 320-1 among all collective variables r arranged in the structure space 300 is the existence ratio ρ ₁ of cluster 310-1. The existence ratios ρ ₂ and ρ ₃ can be calculated in the same manner for clusters 320-2 and 320-3. The stability score T _cc is the sum of the existence ratios ρ ₁ to ρ ₃ .

The stability score _Tcc has a value between 0 and 1. The larger the stability score _Tcc , the greater the fluctuation range of the structure. On the other hand, the smaller the stability score _Tcc , the smaller the fluctuation range of the structure. Therefore, the closer the stability score _Tcc is to 1, the more stable the three-dimensional structure of the compound is.

<Validity of Stability Score _Tcc >
The results of verifying the validity of the stability score _Tcc will be described with reference to Fig. 5 to Fig. 8. In the verification, molecular simulation was performed on two different compounds, and the relationship between the structural fluctuation range and the stability score _Tcc was analyzed.

Figure 5 shows a first example of the root mean square deviation. Figure 6 shows a second example of the root mean square deviation. Figures 5 and 6 are graphs showing the time change in the root mean square deviation (RMSD) calculated for each sample obtained by sampling the three-dimensional structure of one molecule of each different compound at a predetermined time interval. The horizontal axis of the graph is time (in nanoseconds) and the vertical axis is RMSD (in angstroms).

The mean square deviation shown in Figure 5 has an almost constant value. This indicates that very small structural fluctuations occur around the stable structure in the solvent. On the other hand, the mean square deviation shown in Figure 6 has a large range of values, which indicates that large structural fluctuations occur. The mean square deviation shown in Figure 6 also indicates that the structure fluctuates while transitioning between multiple structures.

Figure 7 is a diagram showing a first example of a stability score. The stability score shown in Figure 7 is an example of a stability score calculated from the mean squared deviation shown in Figure 5. Figure 8 is a diagram showing a second example of a stability score. The stability score shown in Figure 8 is an example of a stability score calculated from the mean squared deviation shown in Figure 6.

In the example shown in Figure 7, one cluster (#1) was formed by clustering, and the stability score _Tcc of cluster #1 was 1.00. Therefore, it can be seen that a compound having a mean square deviation as shown in Figure 5 has a very stable structure in a solvent.

In the example shown in Figure 8, four clusters (#1 to #4) were formed by clustering, and the sum of the stability scores _Tcc of clusters #1 to #4 was 0.40. Therefore, it is understood that the compound having the mean square deviation shown in Figure 6 has an unstable structure in the solvent.

The verification results shown in FIG. 5 to FIG. 8 demonstrate that the stability score _Tcc is an index representing the conformational characteristics (stability) of a compound in a solvent.

<Evaluation method processing procedure>
The evaluation method executed by the molecular structure evaluation system 1 in this embodiment will be described with reference to Fig. 9. Fig. 9 is a flow chart showing an example of the evaluation method in this embodiment.

In step S1, the user terminal 20 generates a structure of a modified compound in which a part of the structure of the hit compound is modified in response to the user's operation. The structure of the modified compound can be generated, for example, using known molecular modeling software. In addition, the user terminal 20 determines the environment (e.g., solvent molecules) in which the physical properties of the modified compound will be evaluated in response to the user's operation.

The user terminal 20 generates structural information based on information indicating the structure of the modified compound and information regarding the environment in which the physical properties of the modified compound are evaluated. The user terminal 20 then transmits the structural information to the evaluation device 10.

In the evaluation device 10, the input unit 11 receives structural information from the user terminal 20. The input unit 11 accepts the received structural information. Then, the input unit 11 sends the structural information to the structural analysis unit 12.

In step S2, the structure analysis unit 12 of the evaluation device 10 receives structure information from the input unit 11. Next, the structure analysis unit 12 calculates the three-dimensional structure of the modified compound in a specified environment using a molecular simulation based on molecular dynamics. Next, the structure analysis unit 12 samples the three-dimensional structure of the modified compound generated by the molecular simulation at specified time intervals within a specified time range. Then, the structure analysis unit 12 sends the group of three-dimensional structures obtained by sampling to the collective variable calculation unit 13.

In step S3, the collective variable calculation unit 13 of the evaluation device 10 receives the group of three-dimensional structures from the structure analysis unit 12. Next, the collective variable calculation unit 13 calculates predetermined collective variables for each three-dimensional structure. Then, the collective variable calculation unit 13 places the calculated collective variables in the structure space. This generates a distribution of collective variables for the modified compound. Then, the collective variable calculation unit 13 sends the distribution of collective variables to the cluster division unit 14.

In step S4, the cluster division unit 14 of the evaluation device 10 receives the distribution of the collective variables from the collective variable calculation unit 13. Next, the cluster division unit 14 divides the distribution of the collective variables into one or more clusters. As a result, one or more clusters are formed in the structural space. Then, the cluster division unit 14 sends the structural space in which the clusters are formed to the representative structure determination unit 15.

In step S5, the representative structure determination unit 15 of the evaluation device 10 receives the structural space from the cluster division unit 14. Next, the representative structure determination unit 15 divides the space included in each cluster of the structural space into subspaces. Next, the representative structure determination unit 15 calculates the density of each subspace.

Furthermore, the representative structure determination unit 15 identifies a three-dimensional structure corresponding to the collective variable located closest to the center in the subspace with the highest density. Next, the representative structure determination unit 15 determines the identified three-dimensional structure as a cluster center structure. Then, the representative structure determination unit 15 sends the structure space in which the cluster center structure is set for each cluster to the score calculation unit 16.

In step S6, the score calculation unit 16 of the evaluation device 10 receives the structure space from the representative structure determination unit 15. Next, the score calculation unit 16 calculates a stability score _Tcc based on the structure space in which the cluster center structure is set. Then, the score calculation unit 16 sends the stability score _Tcc to the output unit 17.

In step S7, the output unit 17 of the evaluation device 10 receives the stability score _Tcc from the score calculation unit 16. Next, the output unit 17 generates an evaluation result including at least the stability score _Tcc . Then, the output unit 17 transmits the evaluation result to the user terminal 20.

The user terminal 20 receives the evaluation results from the evaluation device 10. The user terminal 20 then displays the received evaluation results on the display device 506.

The user can evaluate the stability of the modified compound by referring to the evaluation results displayed on the display device 506 of the user terminal 20. In addition, the user can predict the physical properties of the modified compound based on the evaluation results and use them in searching for lead compounds.

<Application Examples>
The stability score _Tcc can be applied to predict the physical properties of modified compounds. For example, based on the stability score _Tcc of a hit compound for a target molecule, it is possible to predict the increase or decrease in binding activity, membrane permeability, heat resistance, etc., due to modification of the hit compound. In this application example, the prediction of the increase or decrease in binding activity will be described. Predictions of other physical properties can also be made in the same manner.

The prediction of an increase or decrease in binding activity will be explained with reference to FIG. 10. FIG. 10 is a diagram showing an example of a prediction of an increase or decrease in binding activity.

10 shows the results of performing molecular simulations on modified compounds and calculating the stability score _Tcc with respect to the cluster center structure of a hit compound for the resulting 3D structure group. The stability score _Tcc is an index showing the similarity to the 3D structure of the hit compound. The closer the stability score _Tcc is to 1, the higher the similarity to the hit compound, and the closer it is to 0, the lower the similarity to the hit compound. Since hit compounds have high binding activity to target molecules, it is expected that there is a correlation between the stability score _Tcc and the increase or decrease in binding activity to the target molecule.

For the hit compound of the cyclic peptide for the target protein, three modified compounds A, B, and C were prepared, and the stability score _Tcc was calculated. Then, the modified compounds A, B, and C were actually synthesized, and the binding activity value Kd was measured by an experiment. FIG. 10 is a graph plotting the data of the hit compound and the modified compounds (A, B, and C) with the binding activity value Kd (nM) on the horizontal axis and the stability score _Tcc (×10 ² ) on the vertical axis. As shown in FIG. 10, a high correlation was observed between the stability score _Tcc and the binding activity value Kd.

The results shown in FIG. 10 demonstrate that it is possible to predict the tendency of increased or decreased binding activity of modified compounds based on the stability score _Tcc of hit compounds.

The membrane permeability of the modified compound can be predicted as follows. First, the stability score _Tcc of the modified compound is calculated in a solvent equivalent to a cell membrane (e.g., chloroform). Next, the stability score Tcc of the modified compound in water is calculated with reference to the cluster center structure determined from the group of three-dimensional structures in the solvent equivalent to the cell membrane. If the stability score _Tcc in water is 0, it can be predicted that the modified _{compound will not permeate the membrane. On the other hand, if the stability score Tcc in} water is finite, it can be predicted that the modified compound will permeate the membrane.

The thermal resistance of the modified compound can be predicted as follows. First, the conformation of the modified compound is sampled by molecular simulation using the replica exchange method. Next, the stability score _Tcc is calculated based on the sampled conformation group. The thermal resistance of the modified compound can be predicted by estimating the temperature at which the stability score _Tcc changes significantly.

Effects of the embodiment
The evaluation device 10 in this embodiment divides the distribution of population variables calculated from the structure of a compound in a predetermined environment into one or more clusters, determines a cluster center structure that represents the cluster from the structures contained in each cluster, and calculates a score based on the proportion of structures present within a predetermined distance from the cluster center structure among the structures contained in the distribution. Therefore, the evaluation device 10 in this embodiment can efficiently evaluate the structure of a compound.

The evaluation device 10 in this embodiment calculates a score for a compound obtained by modifying a compound with known physical properties. Since the score indicates the stability of the compound's structure, the physical properties of the compound are expected to be similar to the known properties. Therefore, the evaluation device 10 in this embodiment can efficiently evaluate the physical properties of a compound.

The evaluation device 10 in this embodiment calculates a score for a compound obtained by modifying a compound whose binding activity to a target substance, membrane permeability, or heat resistance is known. Binding activity, membrane permeability, and heat resistance are important physical properties for a drug candidate compound. Therefore, the evaluation device 10 in this embodiment can calculate a score for efficiently searching for drug candidate compounds.

The stability score _Tcc in this embodiment is an index representing the characteristics of a molecule. A large value indicates that the molecule is stable in a solvent and has a small range of fluctuation in the structure of the system, whereas a small value indicates that the molecule is characterized by a plurality of stable structure groups and has a large range of fluctuation in the structure of the system.

In this embodiment, the stability score _Tcc represents the structural characteristics of a hit compound, and the structural similarity of a modified compound designed from the hit compound to the hit compound can be quantified. If the stability score _Tcc is large, the binding activity to the target molecule increases. On the other hand, if the stability score _Tcc is small, the binding activity to the target molecule decreases.

<Effectiveness in business situations>
The molecular structure evaluation system 1 in this embodiment can be used, for example, in the field of drug discovery. In the field of drug discovery, the enormous cost of developing new drugs is an issue. One of the reasons for this is that experiments or simulations with high computational costs are required to search for lead compounds from hit compounds, which results in high time and financial costs.

The molecular structure evaluation system 1 in this embodiment can efficiently evaluate the physical properties of compounds. Therefore, the molecular structure evaluation system 1 in this embodiment can reduce the time and financial costs required to search for lead compounds. As a result, it becomes possible to develop new and useful drugs at low cost and with a short lead time.

付記１から付記３のいずれかに記載の評価プログラム。
（付記５）
前記化合物は、物性が既知の化合物を改変した化合物である、
付記１から付記４のいずれかに記載の評価プログラム。
（付記６）
前記物性は、標的物質への結合活性、膜透過性及び熱耐性のいずれかを含む、
付記５に記載の評価プログラム。
（付記７）
前記集団変数は、前記構造を示す指標である、
付記１から付記６のいずれかに記載の評価プログラム。
（付記８）
前記集団変数は、前記化合物中の原子位置の平均二乗偏差である、
付記７に記載の評価プログラム。
（付記９）
所定環境における化合物の構造から計算される集団変数の分布を１以上のクラスタに分割する分割部と、
前記クラスタそれぞれに含まれる前記構造から、前記クラスタを代表する代表構造を決定する決定部と、
前記分布に含まれる前記構造のうち前記代表構造から所定距離の範囲に存在する前記構造の割合に基づくスコアを計算する計算部と、
を備える評価装置。
（付記１０）
所定環境における化合物の構造から計算される集団変数の分布を１以上のクラスタに分割し、
前記クラスタそれぞれに含まれる前記構造から、前記クラスタを代表する代表構造を決定し、
前記分布に含まれる前記構造のうち前記代表構造から所定距離の範囲に存在する前記構造の割合に基づくスコアを計算する、
処理をコンピュータが実行する評価方法。 [Additional Notes]
The following supplementary notes are further disclosed regarding the above embodiment.
(Appendix 1)
Dividing a distribution of population variables calculated from the structures of compounds in a given environment into one or more clusters;
determining a representative structure that represents each of the clusters from the structures contained in each of the clusters;
calculating a score based on a ratio of the structures present within a predetermined distance from the representative structure among the structures included in the distribution;
An evaluation program for causing a computer to execute processing.
(Appendix 2)
The score is the sum of the proportions for each of the clusters.
2. The evaluation program according to claim 1.
(Appendix 3)
determining the representative structure based on a density of the structure within the cluster;
3. The evaluation program according to claim 1 or 2.
(Appendix 4)
Let r be the group variable, i be the cluster number, _rc be the distance, _ρi be the proportion, and _cci ^ref be the representative structure, and calculate the score by the following formula:

4. The evaluation program according to claim 1,
(Appendix 5)
The compound is a compound obtained by modifying a compound whose physical properties are known.
5. The evaluation program according to any one of claims 1 to 4.
(Appendix 6)
The physical property includes any one of binding activity to a target substance, membrane permeability, and heat resistance.
6. The evaluation program according to claim 5.
(Appendix 7)
The group variables are indicative of the structure.
7. The evaluation program according to any one of claims 1 to 6.
(Appendix 8)
The population variable is the mean square deviation of the atomic positions in the compound.
8. The evaluation program according to claim 7.
(Appendix 9)
a partitioning unit for partitioning a distribution of population variables calculated from the structures of compounds in a given environment into one or more clusters;
a determination unit that determines a representative structure that represents each cluster from the structures included in each cluster;
a calculation unit that calculates a score based on a ratio of the structures present within a predetermined distance from the representative structure among the structures included in the distribution;
An evaluation device comprising:
(Appendix 10)
Dividing a distribution of population variables calculated from the structures of compounds in a given environment into one or more clusters;
determining a representative structure that represents each of the clusters from the structures contained in each of the clusters;
calculating a score based on a ratio of the structures present within a predetermined distance from the representative structure among the structures included in the distribution;
An evaluation method in which processing is performed by a computer.

The present invention is not limited to the configurations shown here, including combinations of the configurations and other elements in the above-mentioned embodiments. These points can be changed without departing from the spirit of the present invention, and can be appropriately determined according to the application form.

Reference Signs List 1 Molecular structure evaluation system 10 Evaluation device 11 Input section 12 Structure analysis section 13 Population variable calculation section 14 Cluster division section 15 Representative structure determination section 16 Score calculation section 17 Output section 20 User terminal

Claims (10)

Dividing a distribution of population variables calculated from the structures of compounds in a given environment into one or more clusters;
determining a representative structure that represents each of the clusters from the structures contained in each of the clusters;
calculating a score based on a ratio of the structures present within a predetermined distance from the representative structure among the structures included in the distribution;
An evaluation program for causing a computer to execute processing. The score is the sum of the proportions for each of the clusters.
The evaluation program according to claim 1 . determining the representative structure based on density within the cluster;
The evaluation program according to claim 1 . Let r be the group variable, i be the cluster number, _rc be the distance, _ρi be the proportion, and _cci ^ref be the representative structure, and calculate the score by the following formula:

The evaluation program according to claim 1 . The compound is a compound obtained by modifying a compound whose physical properties are known.
The evaluation program according to any one of claims 1 to 4. The physical property includes any one of binding activity to a target substance, membrane permeability, and heat resistance.
The evaluation program according to claim 5 . The group variables are indicative of the structure.
The evaluation program according to any one of claims 1 to 4. The population variable is the mean square deviation of the atomic positions in the compound.
The evaluation program according to claim 7. a partitioning unit for partitioning a distribution of population variables calculated from the structures of compounds in a given environment into one or more clusters;
a determination unit that determines a representative structure that represents each cluster from the structures included in each cluster;
a calculation unit that calculates a score based on a ratio of the structures present within a predetermined distance from the representative structure among the structures included in the distribution;
An evaluation device comprising: Dividing a distribution of population variables calculated from the structures of compounds in a given environment into one or more clusters;
determining a representative structure that represents each of the clusters from the structures contained in each of the clusters;
calculating a score based on a ratio of the structures present within a predetermined distance from the representative structure among the structures included in the distribution;
An evaluation method in which processing is performed by a computer.

Images