JP2019020791A - Toxicity predicting method and utilization thereof - Google Patents

Abstract

To provide compound toxicity predicting means which enables attainment of prediction result which is highly accurate and reliable, and easy in evaluation thereof.SOLUTION: A method for predicting toxicity of a compound executes the steps of: (1) receiving structure information of a test compound inputted by a user; (2) generating a three-dimensional molecular structure in which the structure is optimized, based on the received structure information; (3) calculating values of one or more molecular descriptions including at least one molecular description selected from the group consisting of a three-dimensional molecular description, a four-dimensional molecular description and a quantum chemical molecular description, by using the three-dimensional molecular structure; (4) calculating by a toxicity prediction model a probability as to the presence or absence of toxicity of the test compound by using the value of the molecular description, in which 100% is obtained by summing up the probability with the toxicity and the probability without the toxicity; and (5) outputting the calculated probability.SELECTED DRAWING: Figure 1

Description

  The present invention relates to the prediction of compound toxicity. Specifically, the present invention relates to a method, a system, and a program for predicting the toxicity of a compound.

  The toxicity of a compound is evaluated based on various toxicity indicators (for example, hERG inhibition, biodegradability, mutagenicity) by in vitro and in vivo tests. Each toxicity index has its own determination criteria, and the presence or absence of toxicity is determined according to the determination criteria. Since tests for toxicity evaluation take a lot of time and money, it is desirable to predict the toxicity in advance and select (narrow down) the compounds (candidate compounds) to be used for the test in advance. That is, there is a need to predict the toxicity of a compound without performing actual testing. If toxicity can be predicted in advance, the time and cost required for the test can be reduced as the number of candidate compounds decreases. In addition, it is possible to grasp the toxicity of a compound that cannot be actually tested or is difficult, such as a virtual compound. This advantage is particularly important in the development of new compounds, which increases the design efficiency of new compounds and contributes to improving the success rate and reducing development costs. In response to global trends in animal experiment control in compound development (REACH regulations), the demand for compound toxicity prediction is increasing.

  In the toxicity prediction system / program and the like developed so far, generally, the determination result that the test compound is toxic or not is output together with the prediction accuracy (for example, Patent Documents 1 to 3 and Non-Patent Documents 1 to 5). See). As the prediction accuracy, a matching rate in cross validation or external validation is often used. The higher the match rate value, the higher the prediction accuracy. The coincidence rate is defined as (the number of coincidence of the compound toxicity prediction result and the compound toxicity test result) / (total number of compounds for which the toxicity was predicted). In cross-validation, data is divided into a training set and a test set, a prediction method is constructed using the training set, and the prediction accuracy of the constructed prediction method is verified using test data. In external verification, data independent of the data used for cross-validation is used to verify the prediction accuracy of the constructed prediction method. Even if cross-validation or external verification is used, qualitative determination (that is, toxicity is present or not) is not changed, and comparison between compounds (determination of superiority or inferiority) is difficult. Further, since the determination is not based on the relationship with the structure of the compound, the reliability of the determination result cannot be said to be high.

International Publication No. 2009/025045 Pamphlet International Publication No. 2009/078096 Pamphlet International Publication No. 2010/016109 Pamphlet

Wang S., ET AL, "Recent developments in computational prediction of HERG blockage", Current Topics in Medicinal Chemistry, (The United Arab Emirates), Bentham Science Publishers, 2013, vol. 13, iss. 11, p. 1317-1326 , DOI: 10.2174 / 15680266113139990036 Blay V., ET AL, "Biodegradability Prediction of Fragrant Molecules by Molecular Topology", ACS Sustainable Chemistry and Engineering, (The United States of America), The American Chemical Society Publications, June 2016, vol. 4, iss. 8, p 4224-4231, DOI: 10.1021 / acssuschemeng.6b00717 Jolly R., ET AL, "An evaluation of in-house and off-the-shelf in silico models: Implications on guidance for mutagenicity assessment", Regulatory Toxicology and Pharmacology, (The Netherlands), Elsevier BV, April 2015, vol. 71, iss. 3, p. 388-397, DOI: 10.1016 / j.yrtph.2015.01.010 Greene N., ET AL, "A practical application of two in silico systems for identification of potentially mutagenic impurities", Regulatory Toxicology and Pharmacology, (The Netherlands), Elsevier BV, May 2015, vol. 72, iss. 2, p. 335-349, DOI: 10.1016 / j.yrtph.2015.05.008 Ferrari T., ET AL, "An open source multistep model to predict mutagenicity from statistical analysis and relevant structural alerts", Chemistry Central Journal, (The United Kingdom), Springer Open, July 2010, vol. 4, Suppl. 1, S2 , DOI: 10.1186 / 1752-153X-4-S1-S2 Lazar, in silico toxicology gmbh, website https://lazar.in-silico.ch/predict PASS, Vladimir Poroikov ET AL. Website http://www.pharmaexpert.ru/passonline/ HazardExpert Pro, CompuDrug International, Inc., website http://www.compudrug.com/hazardexpertpro CompuDrug International, Inc., website http://www.compudrug.com/faq

  By the way, a method / system for determining the toxicity of a compound using structural information has also been developed. One of the characteristics of the compound toxicity prediction software Lazar (Non-patent Document 6), which is one of them, is to calculate and display the probability of having toxicity and the probability of having no toxicity with respect to the structural information of the compound inputted by the user. . However, even if the probability of having toxicity and the probability of having no toxicity are added, the probability is not 100%, and it is difficult to evaluate the prediction result. In particular, comparison between compounds is difficult. Another compound toxicity prediction software PASS (Non-Patent Document 7) has the same problem as Laser. Software (HazardExpert Pro) (Non-Patent Document 8) has been developed in which the sum of the probability of toxicity and the probability of non-toxicity in the prediction result is 100%. This software uses the structural information of the compound entered by the user, calculates the probability of toxicity by paying attention to the toxic fragment structure, and displays it. However, as CompuDrug International, Inc., who developed HazardExpert Pro, admits that “the probability of being toxic is not an accurate value” (Non-patent Document 9), its accuracy and reliability are not high.

  Therefore, an object of the present invention is to provide a compound toxicity prediction means that can obtain a prediction result that is highly accurate and reliable and that can be easily evaluated.

In order to solve the above problems, the following inventions are provided.
[1] (1) receiving the structural information of the test compound input by the user;
(2) generating a three-dimensional molecular structure having an optimized structure based on the received structural information;
(3) One or more molecular descriptions using the three-dimensional molecular structure and including at least one molecular descriptor selected from the group consisting of a three-dimensional molecular descriptor, a four-dimensional molecular descriptor, and a quantum chemical molecular descriptor. Generating a child value;
(4) A step of calculating a toxicity prediction model by using the value of the molecular descriptor to determine whether the test compound is toxic or not. When the probability of toxicity and the probability of non-toxicity are added, 100% And (5) outputting the calculated probability,
A method for predicting the toxicity of a compound, comprising:
[2] The prediction method according to [1], wherein step (4) includes the following steps.
(4-1) normalizing the value of the molecular descriptor, and (4-2) calculating the probability of the presence or absence of toxicity of the test compound using the normalized value.
[3] The three-dimensional molecular structure is a three-dimensional molecular structure optimized by a semi-empirical molecular orbital method, a three-dimensional molecular structure optimized by a non-empirical molecular orbital method, or a density functional method. 3D molecular structure whose structure is optimized by 3D molecular structure, molecular dynamics method, semi-empirical molecular orbital method, non-empirical molecular orbital method, or 3D molecular structure and molecular dynamics method searched by density functional method One or more three-dimensional molecular structures selected from the group consisting of three-dimensional molecular structures whose structures are optimized by any combination of semi-empirical molecular orbital methods, non-empirical molecular orbital methods and density functional methods The prediction method according to [1] or [2].
[4] The prediction method according to [1] or [2], wherein the three-dimensional molecular structure is two or more three-dimensional molecular structures whose structures are optimized by a semi-empirical molecular orbital method.
[5] The one or more molecular descriptors according to any one of [1] to [4], including one or more three-dimensional molecular descriptors and one or more quantum chemical molecular descriptors. Prediction method.
[6] The one or more molecular descriptors are one or more three-dimensional molecular descriptors, one or more quantum chemical molecular descriptors, one or more two-dimensional molecular descriptors, one or more one-dimensional molecules. The prediction method according to any one of [1] to [4], including a descriptor and one or more zero-dimensional molecular descriptors.
[7] Any one of [1] to [6], wherein the toxicity prediction model is a toxicity prediction model constructed by machine learning using normalized molecular descriptor values of a plurality of compounds whose presence or absence of toxicity is known. The prediction method according to claim 1.
[8] The prediction method according to [7], wherein the machine learning is at least one machine learning selected from the group consisting of a support vector machine, a Bayesian network, a neural network, Adaboost, random forest, and active learning.
[9] The method further includes the step of generating a chemical formula of the test compound,
In the step (5), the prediction method according to any one of [1] to [8], wherein the generated chemical formula and the probability are output in association with each other.
[10] There are two or more test compounds,
In the step (5), the prediction method according to any one of [1] to [9], wherein the probability is output for each test compound.
[11] The prediction method according to any one of [1] to [10], wherein in step (5), the determination result of the presence or absence of toxicity of the test compound is output together with the probability.
[12] The prediction method according to any one of [1] to [11], wherein the output in step (5) is a display in a tabular format.
[13] The prediction method according to any one of [1] to [12], wherein the toxicity is mutagenicity determined by a reverse mutation test using bacteria.
[14] An input means for inputting structure information of the test compound;
A receiving means for receiving structural information of the test compound input by the user;
First generation means for generating a three-dimensional molecular structure having an optimized structure based on the received structure information;
One or more molecular descriptor values including at least one molecular descriptor selected from the group consisting of a three-dimensional molecular descriptor, a four-dimensional molecular descriptor, and a quantum chemical molecular descriptor using the three-dimensional molecular structure. First calculating means for calculating
A calculation means for the toxicity prediction model to calculate the probability of the presence or absence of toxicity of the test compound using the value of the molecular descriptor, and the sum of the probability of toxicity and the probability of non-toxicity is 100%. A second calculation means, and an output means for outputting the calculated probability;
A system for predicting the toxicity of a compound.
[15] An input device that functions as the input means;
An arithmetic unit that functions as the first generation unit, the first calculation unit, and the second calculation unit;
An output device functioning as the output means;
A main storage device and a control device for controlling the system;
The system according to [14], including:
[16] The system according to [15], further including an auxiliary storage device in which the program is stored.
[17] A process of receiving structural information of the test compound input by the user;
Based on the received structural information, a process for generating a three-dimensional molecular structure with an optimized structure;
One or more molecular descriptor values including at least one molecular descriptor selected from the group consisting of a three-dimensional molecular descriptor, a four-dimensional molecular descriptor, and a quantum chemical molecular descriptor using the three-dimensional molecular structure. A process of calculating
A process for calculating the probability of the presence or absence of toxicity of the test compound using the value of the molecular descriptor, and a process that is 100% when the probability of toxicity and the probability of non-toxicity are added together And a process for outputting the calculated probability;
A program that causes a computer to execute.
[18] A computer-readable storage medium storing the program according to [17].

  According to the present invention, a prediction result with high accuracy and reliability is output by subjecting the structural information of a compound input by a user (user) to a specific process. In the prediction result, the probability of toxicity is 100% when the probability of toxicity is added to the probability of non-toxicity. Therefore, it is easy to evaluate the prediction result, that is, the comparison between the test compounds is easy. For example, if you want a non-toxic compound, implement the present invention, a compound with a 54% probability of non-toxicity (in other words, a 46% probability of toxicity) and a 63% probability of non-toxicity (in other words, toxicity) If a compound with a certain probability of 37%) is found, the latter compound can be selected as a strong candidate by simply comparing the numerical values.

The flowchart of the toxicity prediction method of this invention. Configuration example of toxicity prediction system. The output example of the Ames mutagenicity prediction result of Example 1. The Ames mutagenicity prediction result of Example 2. Structural information in SMILES format of three pesticides (anthraquinone, diquat and chlormequat) whose Ames mutagenicity was predicted in Example 3. Anthraquinone 3D molecular structure. Displayed using internal coordinates. 3D molecular structure of diquat. Displayed using internal coordinates. 3D molecular structure of chlormequat. Displayed using internal coordinates. Molecular descriptor (partial) values used in Example 3. Ames mutagenicity prediction results of three pesticides (anthraquinone, diquat and chlormequat). The Ames mutagenicity prediction result of Comparative Examples 1 and 2.

1. Method for Predicting Toxicity of Compound A first aspect of the present invention relates to a method for predicting toxicity of a compound (hereinafter also referred to as “prediction method of the present invention”). The prediction method of the present invention can evaluate the toxicity of a test compound (a compound whose toxicity is evaluated) without using cells or animals. By using the present invention in combination with toxicity evaluation using cells and animals, extremely efficient toxicity evaluation becomes possible.

  In general, "toxicity of compounds" refers to acute toxicity (oral), acute toxicity (dermal), acute toxicity (inhalation), skin corrosiveness, skin irritation, eye damage / irritation, genotoxicity, carcinogenicity, reproduction It is evaluated by indicators such as toxicity, neurotoxicity, respiratory sensitization, skin sensitization, germ cell mutagenicity, ecotoxicity, bioconcentration, biodegradability. The evaluation index defining “toxicity of the compound” in the prediction method of the present invention is not particularly limited. In a preferred embodiment, the prediction method of the present invention is applied to prediction of toxicity using as an index the mutagenicity determined by a reverse mutation test (bacterial reverse mutation test (Ames test)) using bacteria. The Ames test method and judgment rules are defined in OECD TG 471.

In the present invention, the following steps (1) to (5) are performed.
(1) Step of receiving structural information of the test compound input by the user (2) Step of generating a three-dimensional molecular structure with an optimized structure based on the received structural information (3) The three-dimensional molecule Calculating a value of one or more molecular descriptors including at least one molecular descriptor selected from the group consisting of a three-dimensional molecular descriptor, a four-dimensional molecular descriptor, and a quantum chemical molecular descriptor using a structure; (4) A step of calculating a toxicity prediction model by using the value of the molecular descriptor to determine whether the test compound is toxic or not. When the probability of toxicity and the probability of non-toxicity are added, 100% Step (5) A step of outputting the calculated probability

  Hereinafter, the details of the prediction method of the present invention will be described with reference to the flowchart shown in FIG. The prediction method of the present invention can be executed by a toxicity prediction system described later.

  First, the structural information of the test compound input by the user (user) is received (step (1)). The user prepares structural information of the test compound. The format of the structure information is, for example, SMILES (may be abbreviated as smi format) (references 1 and 2), MDL MOL (reference 3), SDF (reference 3), CDX (binary file created) by PerkinElmer, Inc.'s software ChemDraw (registered trademark), ChemBioDraw (registered trademark)). The number of test compounds is one or more, and in the latter case, structural information is input for each test compound.

  As the test compound, an organic compound having a molecular weight of 800 or less is suitable instead of a polymer compound such as protein, antibody, long-chain DNA or RNA, polystyrene, or polyacrylate. Moreover, it is good to use compounds other than a heavy metal as a test compound.

Next, based on the received structure information, a three-dimensional molecular structure with an optimized structure is generated (step (2)). For example, a semi-empirical molecular orbital method, a non-empirical molecular orbital method, or a density functional method can be used for the structure optimization. Alternatively, the structure may be optimized by searching for a conformation by a molecular mechanics method, a semi-empirical molecular orbital method, a non-empirical molecular orbital method, or a density functional method. In this step, one or more three-dimensional structures with optimized structures are generated by using the above optimization methods alone or in combination of two or more kinds. An example in which the structure is optimized by using the molecular mechanics method, the semiempirical molecular orbital method, the ab initio molecular orbital method and the density functional method is shown below. In determining whether to use two or more types of structure optimization methods together, it is preferable to consider the length of processing time and the load applied to each device (arithmetic device, control device, main storage device, etc.).
<Example 1 of combined use>
First, the structure is optimized by a semi-empirical molecular orbital method, and the obtained three-dimensional molecular structure is further optimized by a non-empirical molecular orbital method or a density functional method.
<Combination example 2>
First, the structure is optimized by the Hartley-Fock method, and the obtained three-dimensional molecular structure is further optimized by the density functional method, the Moller-Plesset perturbation method, the configuration interaction method, or the cluster expansion method.
<Example 3 of combined use>
First, the semi-empirical molecular orbital method is used to optimize the structure, and the resulting three-dimensional molecular structure is further optimized using the Hartley-Fock method. The structure is further optimized by Plesset perturbation method, configuration interaction method, or cluster expansion method.
<Combined use example 4>
Optimize the structure with the Quantum Mechanics / Molecular Mechanics method or our own N-layered integrated molecular orbital and molecular mechanics method.

  In general, many compounds have a plurality of stable three-dimensional structures depending on conditions. Generating a plurality of (that is, two or more) three-dimensional structures with optimized structures reflects this point and provides more useful prediction results. In a preferred embodiment, two or more three-dimensional molecular structures whose structures are optimized by a semi-empirical molecular orbital method are generated, and the process proceeds to the next step.

  “Structural optimization” means minimizing the energy of a molecule by changing the position of atoms constituting the molecule (reference document 4). The “semi-empirical molecular orbital method” is a method of calculating the energy of the electronic state of a molecule based on an approximate expression of the Hartree-Fock equation using empirical parameters (Reference 5). “Non-empirical molecular orbital method” is a method for calculating the energy of the electronic state of a molecule by the Hartley-Fock method or Moller-Plesset perturbation method, configuration interaction method, cluster expansion method, etc. (Reference 6) . The “density functional method” is a method for calculating the energy of the electronic state of a molecule by a functional of electron density (Reference Document 7). The “molecular mechanics method” is a method for calculating molecular potential energy based on a classical mechanical internuclear potential energy function (Reference Document 8). "Conformation search" is to optimize the structure of each conformation after systematically generating a large number of molecular conformations (Reference 9). The Quantum Mechanics / Molecular Mechanics method and our own N-layered integrated molecular orbital and molecular mechanics method are methods for calculating the molecular energy by mixing (hybridizing) a plurality of the aforementioned methods for calculating the molecular energy.

For example, CORINA Classic (reference 10), SYBYL (registered trademark) -X Suite (reference 11), Open Babel (reference 12), The Chemistry Development Kit (reference 13), RDKit (reference 14), Chem3D ^™ (reference 15), ChemBio3D® (reference 16), MarvinSketch (reference 17), Balloon (reference 18), TINKER (reference) Reference 19), Amber (Reference 20), AmberTools (Reference 21), CHARMM (Reference 22), NAMD (Reference 23), BOSS (Reference 24), VEGA ZZ / VEGA Command line (Reference 25) , GROMOS ^™ (reference 26), GROMACS (reference 27), MOPAC® (reference 28), GAMESS (reference 29), Firefly (reference 30), Gaussian® (reference 31) ), Spartan (reference 32), Q-Chem (reference) Reference 33), HyperChem (reference 34), Molecular Operating Environment (reference 35), BIOVIA (registered trademark) Discovery Studio (reference 36), BIOVIA (registered trademark) Material Studio (reference 37), ConfGen (reference) Reference 38), LigPrep (reference 39), Desmond Molecular Dynamics System (reference 40), Jaguar (reference 41), MacroModel (reference 42), MOLGEN (reference 43), CONFLEX (reference) 44), OMEGA (reference 45), VConf (reference 46), Key3D (reference 47), Molpro (reference 48), Molcas (reference 49), ADF (reference 50), TURBOMOLE (reference 51) ), PQS (reference 52), MPQC (reference 53), Dalton (reference 54), LSDalton (reference 55), COLUMBUS (reference 56), NWChem (reference 57), PSI4 (reference 58) , CFOUR Reference 59), ACES (reference 60), ORCA (reference 61), SMASH (reference 62), ABINIT-MP (reference 63), NTChem (reference 64), PAICS (reference 65), etc. Computer software can be used. By instructing the software as described above from the outside, a three-dimensional molecular structure with an optimized structure may be generated.

  Subsequently, one or more molecular descriptor values are calculated using the optimized three-dimensional molecular structure (step (3)). At least one of the molecular descriptors used is a three-dimensional molecular descriptor, a four-dimensional molecular descriptor or a quantum chemical molecular descriptor. Since the structure is based on the optimized 3D molecular structure, accurate and reliable values of molecular descriptors (3D molecular descriptors, 4D molecular descriptors, quantum chemical molecular descriptors, etc.) are calculated. Therefore, it is possible to output a highly accurate prediction result.

  Preferably, two or more molecular descriptors selected from the group consisting of three-dimensional molecular descriptors, four-dimensional molecular descriptors and quantum chemical molecular descriptors (for example, combined use of three-dimensional molecular descriptors and four-dimensional molecular descriptors) And a combination of a 3D molecular descriptor and a quantum molecule descriptor, a 3D molecular descriptor, a 4D molecular descriptor, and a quantum chemical molecular descriptor). For example, if the 3D molecular descriptor is used alone or in combination with other molecular descriptors (ie, 4D molecular descriptors and / or quantum chemical molecular descriptors), preferably 20 or more types, Preferably, 30 or more types, more preferably 50 or more types of three-dimensional molecular descriptor values are calculated. The upper limit of the type (number) of 3D molecular descriptors for which the value is calculated is not particularly limited. However, considering the length of processing time and the load on each device (arithmetic device, control device, main storage device, etc.), for example, 3,084 types can be set as the upper limit. The same applies to the four-dimensional molecular descriptor, and preferably 200 or more values, more preferably 30 types or more, and even more preferably 50 types or more are calculated (the upper limit is, for example, 6,480 types). The same applies to the quantum chemical molecule descriptor, and preferably three or more values, more preferably five or more types, and even more preferably ten or more values are calculated (upper limit is 171 types, for example).

In addition to 3D molecular descriptors, 4D molecular descriptors and / or quantum chemical molecular descriptors, values such as 2D molecular descriptors, 1D molecular descriptors and 0D molecular descriptors may also be calculated. Good. Even in this case, the combination of molecular descriptors whose values are calculated is not particularly limited. Examples of combinations of molecular descriptors whose values are calculated are as follows.
Example 1) One or more 3D molecular descriptors, 1 or more 4D molecular descriptors, 1 or more quantum chemical molecule descriptors, 1 or more 2D molecular descriptors, 1 or more 1D molecules Descriptors, combinations of one or more zero-dimensional molecular descriptors Example 2) one or more three-dimensional molecular descriptors, one or more quantum chemical molecular descriptors, one or more two-dimensional molecular descriptors, one or more 1-dimensional molecular descriptors and combinations of one or more zero-dimensional molecular descriptors Example 3) One or more three-dimensional molecular descriptors, one or more two-dimensional molecular descriptors, one or more one-dimensional molecular descriptors A combination of one or more zero-dimensional molecular descriptors

  The total number of molecular descriptors whose values are calculated is not particularly limited, but preferably 800 or more, more preferably 1,000 or more, and even more preferably 1,500 or more molecular descriptors are calculated. In principle, more reliable prediction results can be obtained by increasing the number of molecular descriptors. On the other hand, if the number of molecular descriptors is too large, there are harmful effects such as excessive processing time and excessive load on each device (arithmetic device, control device, main storage device, etc.). The total number of descriptors should be in the range of 1,000 to 10,000.

  The “three-dimensional molecular descriptor” is a radial distribution function descriptor, a weighted holistic invariant molecular descriptor, a charged partial surface area, or the like (reference document 66). “4D molecular descriptor” includes Comparative Molecular Fields Analysis, GRID, conformational descriptor, 4D molecular fingerprint, and the like (references 66 and 67). “Quantum chemical molecule descriptor” refers to the highest occupied orbital energy, lowest empty orbital energy, ionization potential, electron affinity, dipole moment, and the like (references 66 and 68). The “two-dimensional molecular descriptor” refers to topological distance matrix descriptors, Zagreb Index descriptors, and the like (reference document 66) such as Walk Count Descriptor and Path Count Descriptor which are graph storage amounts. The “one-dimensional molecular descriptor” is the number of functional groups, the number of fragment structures, molecular fingerprints, and the like (reference document 66). The “0-dimensional molecular descriptor” includes the molecular weight, the number of carbon atoms, the number of single bonds capable of free rotation, and the like (reference document 66).

  For calculating the value of the molecular descriptor, for example, DRAGON (reference 69), CODESSA PRO (reference 70), ADAPT (reference 71), ADMET Predictor (reference 72), CORINA Symphony (reference 73), Pentacle (reference 74), VolSurf + (reference 75), ISIDA Fragmentor (reference 76), JOELib (reference 77), Molconn-Z (reference 78), PowerMV (reference 79), PreADMET (reference 80) ), PaDEL-Descriptor (reference 81), cinfony (reference 82), Chemopy (reference 83), The Chemistry Development Kit (reference 13), RDKit (reference 14), Open Babel (reference 12), ToMoCoMD-CARDD (reference 84), QuaSAR-Descriptor (reference 85), Molecular Operating Environment (reference 35), SYBYL (registered trademark) -X Suite (reference 11), BIOVIA (registered trademark) Discovery Studio (reference) Reference 36), BIOVIA (Registered trademark) Material Studio (reference 37), QikProp (reference 86), Jaguar (reference 41), MacroModel (reference 42), VCharge (reference 87), MarvinSketch (reference 17), Spartan (reference) Reference 32), MOPAC (registered trademark) (reference 28), GAMESS (reference 29), Gaussian (registered trademark) (reference 31), HyperChem (reference 34), Q-Chem (reference 33), BOSS (Reference 24), Firefly (reference 30), Molpro (reference 48), Molcas (reference 49), ADF (reference 50), TURBOMOLE (reference 51), PQS (reference 52), MPQC ( Reference 53), Dalton (reference 54), LSDalton (reference 55), COLUMBUS (reference 56), NWChem (reference 57), PSI4 (reference 58), CFOUR (reference 59), ACES (reference) Reference 60), ORCA (Reference 61), SMAS Computer software such as H (reference 62), ABINIT-MP (reference 63), NTChem (reference 64), PAICS (reference 65), Mold2 (reference 88) can be used. The value of the molecular descriptor may be calculated by instructing the software as described above from the outside.

  Following step (3), the probability of the presence or absence of toxicity of the test compound is calculated (step (4)). A toxicity prediction model is used to calculate the probability. In the toxicity prediction model, the probability of the presence or absence of toxicity of the test compound is calculated using the value of the molecular descriptor calculated in step (3). One of the greatest features of the present invention is to calculate the probability of presence / absence of toxicity so that the probability of toxicity and the probability of non-toxicity are 100%. When there are two or more test compounds, the probability of toxicity is calculated for each test compound.

The toxicity prediction model may be a toxicity prediction model constructed by machine learning using normalized molecular descriptor values of a plurality of compounds with known toxicity. Examples of machine learning are support vector machines, Bayesian networks, neural networks, Adaboost, random forest, active learning. Two or more of these may be used in combination. Machine learning includes, for example, LibSVM (reference 89), TensorFlow ^™ (reference 90), Chainer (registered trademark) (reference 91), Jubatus (registered trademark) (reference 92), and Caffe (reference 93). , Theano (reference 94), Torch (reference 95), neon ^™ (reference 96), MXNet (reference 97), The Microsoft Cognitive Toolkit (reference 98), R (C) (reference 99), MATLAB (registered trademark) (reference 100), Mathematica (registered trademark) (reference 101), SAS (registered trademark) (reference 102), RapidMiner (registered trademark) (reference 103), KNIME (registered trademark) ( Reference 104), WeKa (reference 105), shogun-toolbox / shogun (reference 106), Orange (reference 107), Apache Mahout ^™ (reference 108), scikit-learn (reference 1095), mlpy ( Reference 110), XGBoost (see Document 111), it can be utilized computer software such as Deeplearning4j (ref 112). The machine learning may be executed by instructing the software as described above from the outside.

  Basically, the more types of known compounds that are used to build a toxicity prediction model, the more reliable the toxicity prediction model. Preferably, 300 or more, more preferably 3,000 or more, and even more preferably 7,500 or more known compounds are used in the construction of a toxicity prediction model.

As step (4), the following two steps are preferably performed.
(4-1) Step of normalizing the value of molecular descriptor (4-2) Step of calculating probability of presence or absence of toxicity of test compound using normalized value

  Step (4-1) is a step for enabling comparison with the corresponding molecular descriptors of a plurality of compounds whose presence or absence of toxicity is known. For example, calculation (processing) for normalizing the molecular descriptor values for a plurality of compounds with known toxicity is performed. Using the normalized value obtained in this step, the probability of the presence or absence of toxicity of the test compound is calculated in step (4-2).

The probability calculated in step (4) is output in a predetermined format (step (5)). It is possible to output in various formats. For example, it is displayed in a table format or a graph format. Preferably, it is output so that it can be read / displayed by general-purpose software such as Excel (registered trademark) (reference document 117), Libre Office (reference document 118), Apache Open Office ^™ (reference document 119). When there are two or more test compounds, the probability of the presence or absence of toxicity of each test compound is output, and the typical display mode is a list of the probabilities of all test compounds. It is what is displayed, but it is not limited to this.

  Along with the probability of presence / absence of toxicity, a determination result (typically “toxic”, “non-toxic” or similar) of the test compound may be output. The determination result makes it possible, for example, to more efficiently select or select a candidate compound (a promising compound expected to have low toxicity or no toxicity) from a plurality of test compounds.

In one embodiment of the present invention, a step of generating a chemical formula of the test compound (step (6)) is also performed. In step (5), the chemical formula generated in the step and the probability calculated in step (4) are calculated. Output in association (for example, integrated into a table format). Such an output will show a relationship between the structure of the compound and toxicity, which is a more useful predictive result. As the “chemical formula”, a chemical structural formula, a molecular formula, or the like can be adopted. Preferably, the chemical structural formula is adopted because the user can recognize the geometric structure of the compound. For example, the Chemistry Development Kit (reference 13), RDKit (reference 14), Open Babel (reference 12), MedChem Designer ^™ (reference 113), ChemBioDraw (registered trademark) (reference) 114), ChemDraw (registered trademark) (reference document 115), MarvinSketch (reference document 17), BIOVIA (registered trademark) Draw (reference document 116) and the like can be used. Chemical formulas may be generated by externally instructing the software as described above.

  The prediction results according to the present invention are typically used for the next stage toxicity assessment. Specifically, based on the prediction result of the present invention, a compound to be subjected to toxicity evaluation using cells or animals is selected or selected. Thus, by using the present invention, extremely efficient toxicity evaluation is realized.

2. Toxicity Prediction System FIG. 2 is a diagram conceptually showing a configuration example of the toxicity prediction system of the present invention. The toxicity prediction system 1 in this example is a computer system including an input device 2, an arithmetic device 3, an output device 4, a main storage device 5, a control device 6, and an auxiliary storage device 7. A solid line arrow in FIG. 2 indicates a data flow direction. The broken line arrows in FIG. 2 indicate the flow direction of the control signal. The toxicity prediction system of the present invention can also be constructed using any general-purpose computer.

  The input device 2 is, for example, a keyboard, a mouse, a touch panel, etc., and the user operates the input device to input structural information of one or more test compounds. The main storage device (main memory) 5 is a RAM and / or a ROM. The main storage device 5 captures and stores programs and data stored in the auxiliary storage device 7. The auxiliary storage device 7 is a hard disk drive, an optical disk device, an SSD, or the like. The program may be installed from a computer-readable recording medium or from another computer / server on a network or cloud.

  The control device 6 controls other devices in accordance with a program fetched and stored in the main storage device 5. The auxiliary storage device 7 can store the output of the computer system. The output device 4 is a display, for example. The user can view the output of the computer system via the output device. The arithmetic device 3 takes in the data stored in the main storage device 5, performs an operation based on the operation instruction sent from the control device 6, and returns the operation result to the main storage device 5.

3. Program, Storage Medium The present invention also provides a computer program used for the toxicity prediction system. The computer program of the present invention causes a computer to execute the following processes (i) to (v). The computer program of the present invention includes, for example, CD (Compact Disc) -ROM, CD-R, CD-RW, DVD (Digital Versatile Disc), DVD-RAM, BD (Blu-ray (registered trademark) Disc), Provided in a form stored in a computer-readable storage medium such as MO (Magneto Optical disc), SSD, magnetic tape, or various memory cards (USB flash memory, SD memory card, etc.), or downloaded from a cloud computer, etc. The It is also possible to store the computer program of the present invention in an auxiliary storage device of a computer connected via a network, or to transfer the computer program of the present invention to another computer via the network.
(I) Process for receiving structural information of a test compound input by a user (ii) Process for generating a three-dimensional molecular structure with an optimized structure based on the received structural information (iii) The three-dimensional molecule A process for calculating one or more molecular descriptor values from the structure. (Iv) A process for calculating a probability of the presence or absence of toxicity of the test compound using the molecular descriptor value, wherein the toxicity prediction model calculates the toxicity. Processing that is 100% when the probability of existence and the probability of non-toxicity are added (v) Processing that outputs the calculated probability

<Example 1: Toxicity prediction of two kinds of pesticides>
Outline The toxicity prediction method of the present invention (FIG. 1) was executed using the general-purpose computer system shown in FIG. In order to optimize the structure of the compound, software GAMESS that can execute the pm3 method, which is one of semi-empirical molecular orbital methods, was used. In addition, from the optimized 3D molecular structure, 35 types of 3D molecular descriptors, 4 types of quantum chemical molecular descriptors, 121 types of 0D molecular descriptors, 907 types of 1D molecular descriptors, 2D molecular descriptors We decided to calculate 160 values. Except for some molecular descriptors (calculated with reference to 68 and 120), molecular descriptor values were calculated using the software PaDEL-descriptor (81 and 121). . On the other hand, using the normalized molecular descriptor values of about 8,000 compounds with known Ames mutagenicity, a toxicity prediction model was constructed by machine learning using a support vector machine.

  Structure information of two pesticides (diclolobenil, teflubenzuron) in SMILES format (Clc1cccc (Cl) c1C # N, Fc1cccc (F) c1C (= O) NC (= O) Nc1cc (Cl) c (F) c (Cl) c1F), the chemical structure of the compound and the prediction result were integrated and output in tabular form, and the SMILES structure information is the primary structure of the compound and the format has been established. (Reference Document 1 and Reference Document 2) SMILES format structure files include existing software (for example, MarvinSketch (reference document 17), ChemDraw (registered trademark) (reference document 115), BIOVIA (registered trademark) Draw (reference document 116), etc. ).

  FIG. 3 shows the predicted results of mutagenicity determined by a reverse mutation test using bacteria for two pesticides. The reverse mutation test using bacteria is called the Ames test. Two pesticides used as test compounds are known not to have Ames mutagenicity. As shown in FIG. 3, the probabilities of Ames mutagenicity of the two pesticides were calculated with appropriate and high accuracy. Moreover, since the sum of the probability of having Ames mutagenicity and the probability of having no Ames is 100%, it is easy to compare between compounds.

<Example 2: Toxicity prediction of 724 pesticides (verification of prediction accuracy)>
In order to verify the prediction accuracy of the system of the present invention, Ames mutagenicity of 724 pesticides with known Ames mutagenicity was predicted. As a result, the coincidence rate between the prediction result by the system of the present invention and the result of the Ames test is 657/724 × 100 = 90.7 (%), which proves that the prediction accuracy of the system of the present invention is high (FIG. 4). ). In addition, the prediction system of the present invention can output prediction results for all 724 types. That is, it was shown that it was extremely excellent in practicality.

<Example 3: Comparison of prediction accuracy using three types of pesticides>
Ames mutagenicity of three pesticides (anthraquinone, diquat and chlormequat) was predicted by the method of the present invention. The points not particularly mentioned, such as the method for optimizing the structure, are the same as in the above embodiment.

  FIG. 5 shows the structure information in SMILES format of three pesticides whose Ames mutagenicity was evaluated in this example. In FIGS. 6-8, the three-dimensional molecular structure of three kinds of pesticides generated based on the structural information is displayed using internal coordinates. The internal coordinates consist of a bond length defined at the position of 2 atoms, a bond angle defined at the position of 3 atoms, and a twist angle defined at the position of 4 atoms. The unit of bond length is angstrom. The unit of the bond angle and the twist angle is ° (degrees).

  Based on the 3D molecular structure, the values of 35 types of 3D molecular descriptors, 4 types of quantum chemical molecular descriptors, 119 types of 0D molecular descriptors, 795 types of 1D molecular descriptors, and 149 types of 2D molecular descriptors Was calculated. A part of the calculated values is shown in FIG. To construct the toxicity prediction model, normalized molecular descriptor values of 9,719 compounds with known Ames mutagenicity were used.

  The output of the prediction results for the three types of pesticides is shown in FIG. For each pesticide, the probability of having Ames mutagenicity and the probability of not having Ames mutagenicity (100% by adding the two probabilities) are calculated and output. The three pesticides used as test compounds have been found to be free of Ames mutagenicity by the Ames test.

As a comparative example, the prediction result when the generated molecular descriptor was changed was obtained.
<Comparative Example 1>
The values of 119 kinds of 0-dimensional molecular descriptors, 795 kinds of 1-dimensional molecular descriptors, and 149 kinds of 2-dimensional molecular descriptors will be calculated (excluding 35 kinds of 3-dimensional molecular descriptors and 4 kinds of quantum chemical molecular descriptors). ) Ames mutagenicity was predicted by the same treatment as above. The difference from the above system (Example 3) is that it does not include 35 types of 3D molecular descriptors and 4 types of quantum chemical molecular descriptors.
<Comparative example 2>
Based on the non-optimized 3D molecular structure, 29 types of 3D molecular descriptors, 119 types of 0D molecular descriptors, 795 types of 1D molecular descriptors, and 149 types of 2D molecular descriptors are calculated. In particular, Ames mutagenicity was predicted by the same treatment as above. The difference from the above system (Example 3) is that the value of the 3D molecular descriptor is calculated based on the 3D structure that is not structurally optimized, and does not include the 4 types of quantum chemical molecular descriptors. Is a point.

  The prediction results of Comparative Examples 1 and 2 are shown in FIG. Compared with the output result of FIG. 3 (FIG. 10), it can be seen that both of Comparative Examples 1 and 2 have a high probability of Ames mutagenicity for all three types of pesticides and are inferior in prediction accuracy and accuracy. . It should be noted that in Comparative Examples 1 and 2, the probability of diquat having Ames mutagenicity was higher than the probability without Ames mutagenicity, and the prediction results were contrary to the results of the Ames test.

  According to the present invention, it is possible to predict toxicity of not only known compounds but also compounds to be developed in the future (including virtual compounds). In addition, it is possible to predict the toxicity of a compound that cannot be actually tested or is difficult. Used for chemicals, pharmaceuticals, agricultural chemicals, veterinary drugs (livestock and pets), cosmetics, detergents, dyes, inks, additives, and other materials that require the development of non-toxic or low-toxic compounds. The invention can be applied.

  The present invention is not limited to the description of the embodiments and examples of the invention described above. Various modifications may be included in the present invention as long as those skilled in the art can easily conceive without departing from the description of the scope of claims. The contents of papers, published patent gazettes, patent gazettes, and the like specified in this specification are incorporated by reference in their entirety.

<References>
1. Weininger D., "SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules", Journal of Chemical Information and Modeling, (The United States of America), The American Chemical Society Publications, February 1988, vol. 28, iss.1, p. 31-36, DOI: 10.1021 / ci00057a005
2. Weininger D., ET AL, "SMILES. 2. Algorithm for generation of unique SMILES notation", Journal of Chemical Information and Modeling, (The United States of America), The American Chemical Society Publications, May 1989, vol. 29 , iss. 2, p. 97-101, DOI: 10.1021 / ci00062a008
3. Dalby A., ET AL, "Description of several chemical structure file formats used by computer programs developed at Molecular Design Limited", Journal of Chemical Information and Computer Sciences, (The United States of America), The American Chemical Society Publications, vol. 32, iss. 3, May 1992, p. 244-255. DOI: 10.1021 / ci00007a012
4. Schlegel HB, "Geometry optimization", Wiley Interdisciplinary Reviews: Computational Molecular Science, (The United States of America), John Wiley & Sons, Ltd., May 2011, vol. 1, iss. 5, p.790-809 , DOI: 10.1002 / wcms.34
5. Zerner MC, ET AL (Eds), "Semiempirical Molecular Orbital Methods", Reviews in Computational Chemistry, (Germany), Wiley-VCH, Inc., 1991, vol. 2, p. 313-365, DOI: 10.1002 / 9780470125793.ch8
6. Friesner RA, "Ab initio quantum chemistry: Methodology and applications", Proceedings of the National Academy of Sciences of the United States of America, (The United States of America), The United States National Academy of Sciences, May 2005, vol 102, no. 19, p. 6648-6653, DOI: 10.1073 / pnas.0408036102
7. Parr RG, "Density Functional Theory", Annual Review of Physical Chemistry, (The United States of America), Annual Reviews, Inc., 1983, vol. 34, p. 631-656, https://doi.org /10.1146/annurev.pc.34.100183.003215
8. Bowen JP ET AL, Lipkowitz KB ET AL (Eds), "Molecular Mechanics: The Art and Science of Parameterization", Reviews in Computational Chemistry, (Germany), WILEY-VCH, Inc., 1991, vol. 2, p 81-97, DOI: 10.1002 / 9780470125793
9. Mazzanti A., ET AL, "Recent trends in conformational analysis", Wiley Interdisciplinary Reviews: Computational Molecular Science, (The United States of America), John Wiley & Sons, Inc., November 2011, vol. 2, iss. 4, p. 613-641, DOI: 10.1002 / wcms.96
10. CORINA Classic, Molecular Networks GmbH and Altamira, LLC., Website https://www.mn-am.com/products/corina
11. SYBYL®-X Suite, Certara USA, Inc., website https://www.certara.com/software/molecular-modeling-and-simulation/sybyl-x-suite/
12. O'Boyle NM, ET AL, "Open Babel: An open chemical toolbox", Journal of Cheminformatics, (The United States of America), Springer Publishing, 3:33, Oct 2011, DOI: 10.1186 / 1758-2946- 3-33
13. Steinbeck C., ET AL, "The Chemistry Development Kit (CDK): An Open-Source Java Library for Chemo- and Bioinformatics", Journal of Chemical Information and Computer Sciences, (The United States of America), The American Chemical Society Publications, vol. 43, iss. 2, February 2003, p. 493-500, DOI: 10.1021 / ci025584y
14. RDKit, Landrum G., "RDKit Documentation Release 2017.03.1", Online Documentation, website http://www.rdkit.org/docs/Overview.html
15. Chem3D ^TM , PerkinElmer, Inc., website http://www.cambridgesoft.com/Ensemble_for_Chemistry/ChemOffice/ChemOfficeProfessional/
16. ChemBio3D®, PerkinElmer, Inc., website https://www.cambridgesoft.com/Ensemble_for_Chemistry/details/Default.aspx?fid=17&pid=660
17. MarvinSketch, ChemAxon, website https://www.chemaxon.com/products/calculator-plugins/molecular-modelling/
18. Balloon, Vainio M., website http://users.abo.fi/mivainio/balloon/
19. TINKER, Ponder J., ET AL, website https://dasher.wustl.edu/tinker/
20. Amber, Case DA, ET AL, website http://ambermd.org/
21. AmberTools, website http://ambermd.org/#AmberTools
22. CHARMM, Karplus M., ET AL, website https://www.charmm.org/charmm/
23. NAMD, Theoretical and Computational Biophysics Group, Beckman Institute, University of Illinois, website http://www.ks.uiuc.edu/Research/namd/
24. BOSS, Jorgensen WL, ET AL, website http://zarbi.chem.yale.edu/software.html
25. VEGA ZZ / VEGA Command line, Pedretti A., ET AL: Website http://www.vegazz.net/
26. GROMOS ^TM , van Gunsteren WF, ET AL, website http://www.igc.ethz.ch/gromos.html
27. GROMACS, van der Spoel D., ET AL, website http://www.gromacs.org/
Website http://openmopac.net/
29. GAMESS, Schmidt MW, ET AL, Website http://www.msg.ameslab.gov/gamess/
30. Firefly, Granovsky AA, ET AL, website http://classic.chem.msu.su/gran/gamess/index.html
31. Gaussian (registered trademark), Gaussian, Inc., website http://gaussian.com/citation/
32. Spartan, Wavefunction, Inc., website https://www.wavefun.com/products/spartan.html
33. Q-Chem, Q-Chem, Inc., website http://www.q-chem.com/
34. HyperChem, Hypercube, Inc., website http://www.hyper.com/
35. Molecular Operating Environment, Chemical Computing Group ULC, website https://www.chemcomp.com/MOE-Molecular_Operating_Environment.htm
36. BIOVIA® Discovery Studio, Dassault Systemes, website http://accelrys.com/products/collaborative-science/biovia-discovery-studio/
37. BIOVIA® Material Studio, Dassault Systemes, website http://accelrys.com/products/collaborative-science/biovia-materials-studio/
38. ConfGen, Schrodinger (R), LLC, Schrodinger Release 2017-1: New York, NY, 2017 Website https://www.schrodinger.com/confgen
39. LigPrep, Schrodinger (R), LLC, Schrodinger Release 2017-1: New York, NY, 2017, website https://www.schrodinger.com/ligprep
40. Desmond Molecular Dynamics System, DE Shaw Research, New York, NY, 2017. Website https://www.schrodinger.com/desmond
41. Jaguar, Schrodinger (R), LLC, Schrodinger Release 2017-1: New York, NY, 2017, Website https://www.schrodinger.com/jaguar
42. MacroModel, Schrodinger (R), LLC, Schrodinger Release 2017-1: New York, NY, 2017, Website https://www.schrodinger.com/macromodel
43. MOLGEN, Wassermann A., ET AL, website http://www.molgen.de/
44. CONFLEX (registered trademark), CONFLEX Corporation, website http://www.conflex.net/
45. OMEGA, OpenEye Scientific Software, website https://www.eyesopen.com/omega
46. VConf, VeraChem, LLC, website http://www.verachem.com/products/vconf/
47. Key3D, IMMD, Inc., website http://www.immd.co.jp/en/product_2.html
48. Molpro, TTI GmbH, website https://www.molpro.net/
49. Molcas, Veryazov. V. ET AL, website http://www.molcas.org/
50. ADF (R), Scientific Computing & Modeling NV, Website https://www.scm.com/product/adf/
51. TURBOMOLE, TURBOMOLE GmbH, Website http://www.turbomole.com/
52. PQS, Parallel Quantum Solutions, website http://www.pqs-chem.com/
53. MPQC, Valeev E. ET AL, website http://www.mpqc.org/
54. Dalton, Dalton / LSDalton developers, website http://daltonprogram.org
55. LSDalton, Dalton / LSDalton developers website http://daltonprogram.org
56. COLUMBUS, Lischka H. ET AL, Website https://www.univie.ac.at/columbus/
57. NWChem, Valiev M., ET AL, website http://www.nwchem-sw.org/index.php/Main_Page
58. PSI4, Sherrill CD, ET AL, website http://www.psicode.org/
59. CFOUR, Stanton JF, ET AL, website http://www.cfour.de/
60. ACES, Bartlett RJ, ET AL, Website http://www.qtp.ufl.edu/ACES/
61. ORCA, Neese, F., ET AL, Website https://orcaforum.cec.mpg.de/
62. SMASH, Ishimura K., Website http://smash-qc.sourceforge.net/
63. ABINIT-MP (registered trademark), Mochizuki Y., ET AL, website http://www.cenav.org/abinitmpopen1/
64. NTChem, Nakajima, T., ET AL, Website http://labs.aics.riken.jp/nakajimat_top/ntchem_e.html
65. PAICS, Ishikawa T., Website http://www.paics.net/index_e.html
66. Johann Gasteiger J., ET AL (Editor), "Chemoinformatics: A Textbook", WILEY-VCH Verlag GmbH & Co. KgaA, Weinheim, 2003, ISBN 3-527-30681-1
67. Wicker JGP, ET AL, "Beyond Rotatable Bond Counts: Capturing 3D Conformational Flexibility in a Single Descriptor", Journal of Chemical Information and Modeling, (The United States of America), The American Chemical Society Publications, vol. 56, iss 12, November 2016, p. 2347-2352, DOI: 10.1021 / acs.jcim.6b00565
68. Karelson M., ET AL, "Quantum-Chemical Descriptors in QSAR / QSPR Studies", Chemical Reviews, (The United States of America), The American Chemical Society Publications, vol. 96, iss. 3, May 1996, p 1027-1044, DOI: 10.1021 / cr950202r
69. DRAGON, Kode srl, website https://chm.kode-solutions.net/products_dragon.php
70. CODESSA PRO, CompuDrug International, Inc., website http://www.compudrug.com/codessa_pro
71. ADAPT, Jurs PC, ET AL, website http://research.chem.psu.edu/pcjgroup/adapt.html
72. ADMET Predictor, Simulations Plus, Inc., website http://www.simulations-plus.com/software/admet-property-prediction-qsar/
73. CORINA Symphony, Molecular Networks GmbH and Altamira, LLC., Website https://www.mn-am.com/products/corinasymphony
74. Pentacle, Molecular Discovery Ltd, website http://www.moldiscovery.com/software/pentacle/
75. VolSurf +, Molecular Discovery Ltd, website http://www.moldiscovery.com/software/vsplus/
76. ISIDA Fragmentor, Varnek A., ET AL, website http://infochim.u-strasbg.fr/spip.php?rubrique41
77. JOELib, Zell A., ET AL, website http://www.ra.cs.uni-tuebingen.de/software/joelib/index.html
78. Molconn-Z, eduSoft, LC, website http://www.edusoft-lc.com/molconn/
79. PowerMV, Liu J., ET AL, website https://www.niss.org/research/software/powermv
80. PreADMET, Bioinformatics & Molecular Design Research Center (BMDRC), website https://preadmet.bmdrc.kr/preadmet-pc-version-2-0/
81. PaDEL-Descriptor, Yap CW, Website http://www.yapcwsoft.com/dd/padeldescriptor/
82. Cinfony, O'Boyle NM, ET AL, website http://cinfony.github.io/
83. Cao D.-S., ET AL, "ChemoPy: freely available python package for computational biology and chemoinformatics", Bioinformatics, (The United Kingdom), Oxford University Press, vol. 29, iss. 8, March 2013, p 1092-1094, DOI: https://doi.org/10.1093/bioinformatics/btt105
84. ToMoCoMD-CARDD, Ponce YM, Website http://tomocomd.com/
85. QuaSAR-Descriptor, Chemical Computing Group ULC, website https://www.chemcomp.com/journal/descr.htm
86. QikProp, Schrodinger (R), LLC, Schrodinger Release 2017-1: New York, NY, 2017, Website https://www.schrodinger.com/qikprop
87. VCharge, VeraChem, LLC, website http://www.verachem.com/products/vcharge/
88. Mold2, Hong, H., ET AL, Website https://www.fda.gov/ScienceResearch/BioinformaticsTools/Mold2/ucm144528.htm
89.LibSVM, Chang C.-C., ET AL, website https://www.csie.ntu.edu.tw/~cjlin/libsvm/
90. TensorFlow ^TM , Google Inc., website https://www.tensorflow.org/
91. Chainer (R), Preferred Networks, Inc., website http://chainer.org/
92. Jubatus (registered trademark), Preferred Networks, Inc. and Nippon Telegraph and Telephone Corporation, website https://jubat.us/en/
93. Caffe, Jia Y., ET AL, website http://caffe.berkeleyvision.org/
94. Theano, Theano Development Team, website http://deeplearning.net/software/theano/
95. Torch, Ronan Collobert ET AL, website http://torch.ch/
96.neon ^TM , Nervana Systems, website https://www.nervanasys.com/technology/neon/
97. MXNet, website http://mxnet.io/
98. The Microsoft Cognitive Toolkit, Microsoft Corporation, website https://www.microsoft.com/en-us/cognitive-toolkit/
99. R (c), The R Foundation, website https://www.r-project.org/
100. MATLAB®, The MathWorks, Inc., website https://www.mathworks.com/products/matlab.html?s_tid=hp_products_matlab
101. Mathematica®, Wolfram Research, website http://www.wolfram.com/mathematica/
102. SAS®, SAS Institute Inc., website https://www.sas.com/en_us/home.html
103. RapidMiner (R), RapidMiner, Inc., website https://rapidminer.com/
104. KNIME (registered trademark), KNIME.com AG, website https://www.knime.org/
105. Witten IH, ET AL, The WEKA Workbench. Online Appendix for "Data Mining: Practical Machine Learning Tools and Techniques", (The United States of America), Morgan Kaufmann, Fourth Edition, 2016, ISBN13: 978-0128042915. Site http://www.cs.waikato.ac.nz/ml/weka/
106.shogun-toolbox / shogun, website http://www.shogun-toolbox.org/, Sonnenburg S., ET AL, "shogun-toolbox / shogun: Shogun 6.0.0-Baba Nobuharu", April 2017, DOI : 10.5281 / zenodo.556748
107. Orange, Demsar J., ET AL, "Orange: Data Mining Toolbox in Python", Journal of Machine Learning Research, (The United States of America), JMLR, Inc. and Microtome Publishing, vol. 14, Aug 2013, p. 2349-2353. Website https://orange.biolab.si/
108. Apache Mahout ^TM , The Apache Software Foundation, website http://mahout.apache.org/
109. scikit-learn, website http://scikit-learn.org/stable/. Pedregosa F., ET AL, "Scikit-learn: Machine Learning in Python", Journal of Machine Learning Research, (The United States of America), JMLR, Inc. and Microtome Publishing, vol. 12, Oct 2011, p. 2825-2830
110.mlpy, Albanese D., ET AL, website http://mlpy.sourceforge.net/
111. Chen T., ET AL, "XGBoost: A Scalable Tree Boosting System", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Aug 2016, DOI: 10.1145 / 2939672.2939785
112. Deeplearning4j, Skymind, Deeplearning4j Development Team, "Deeplearning4j: Open-source distributed deep learning for the JVM", Apache Software Foundation License 2.0. Website https://deeplearning4j.org/
113. MedChem Designer ^™ , Simulations Plus, Inc., website http://www.simulations-plus.com/software/medchem-designer/
114. ChemBioDraw (R), PerkinElmer, Inc., website https://www.cambridgesoft.com/Ensemble_for_Chemistry/details/Default.aspx?fid=17&pid=660
115. ChemDraw®, PerkinElmer, Inc., website http://www.cambridgesoft.com/software/overview.aspx
116. BIOVIA® Draw, Dassault Systemes, website http://accelrys.com/products/collaborative-science/biovia-draw/
117. Excel (registered trademark), Microsoft Corporation, website https://products.office.com/en-us/excel
118. Libre Office, The Document Foundation, website https://www.libreoffice.org/
119. Apache Open Office ^TM , The Apache Software Foundation, website https://www.openoffice.org/
120. Sakuratani Y, ET AL, "Molecular size as a limiting characteristic for bioconcentration in fish", Journal of Environmental Biology, January 2008, vol. 29, iss. 1, p. 89-92., Website http: // www.jeb.co.in/journal_issues/200801_jan08/paper_15.pdf
121. Yap CW, "PaDEL-descriptor: An open source software to calculate molecular descriptors and fingerprints", Journal of Computational Chemistry, Volume 32, Issue 7, p. 1466-1474, May 2011. DOI: 10.1002 / jcc.21707. Supporting Information: JCCT_21707_sm_suppinformation.xls, website http://onlinelibrary.wiley.com/doi/10.1002/jcc.21707/suppinfo

DESCRIPTION OF SYMBOLS 1 Toxicity prediction system 2 Input device 3 Arithmetic device 4 Output device 5 Main memory device 6 Control device 7 Auxiliary memory device

Claims (18)

(1) receiving the structural information of the test compound input by the user;
(2) generating a three-dimensional molecular structure having an optimized structure based on the received structural information;
(3) One or more molecular descriptions using the three-dimensional molecular structure and including at least one molecular descriptor selected from the group consisting of a three-dimensional molecular descriptor, a four-dimensional molecular descriptor, and a quantum chemical molecular descriptor. Generating a child value;
(4) A step of calculating a toxicity prediction model by using the value of the molecular descriptor to determine whether the test compound is toxic or not. When the probability of toxicity and the probability of non-toxicity are added, 100% And (5) outputting the calculated probability,
A method for predicting the toxicity of a compound, comprising: The prediction method according to claim 1, wherein step (4) includes the following steps.
(4-1) normalizing the value of the molecular descriptor, and (4-2) calculating the probability of the presence or absence of toxicity of the test compound using the normalized value.   The three-dimensional molecular structure is a three-dimensional molecular structure whose structure is optimized by a semi-empirical molecular orbital method, a three-dimensional molecular structure whose structure is optimized by a non-empirical molecular orbital method, or a structure by a density functional method. Optimized three-dimensional molecular structure and molecular mechanics method, semi-empirical molecular orbital method, non-empirical molecular orbital method or three-dimensional molecular structure conformed by density functional method, molecular mechanics method, semi-empirical One or more three-dimensional molecular structures selected from the group consisting of a three-dimensional molecular structure whose structure is optimized by any combination of the molecular molecular orbital method, the ab initio molecular orbital method and the density functional method Item 3. The prediction method according to item 1 or 2.   The prediction method according to claim 1 or 2, wherein the three-dimensional molecular structure is two or more three-dimensional molecular structures whose structures are optimized by a semi-empirical molecular orbital method.   The prediction method according to any one of claims 1 to 4, wherein the one or more molecular descriptors include one or more three-dimensional molecular descriptors and one or more quantum chemical molecular descriptors.   The one or more molecular descriptors are one or more three-dimensional molecular descriptors, one or more quantum chemical molecular descriptors, one or more two-dimensional molecular descriptors, one or more one-dimensional molecular descriptors, The prediction method according to claim 1, comprising one or more zero-dimensional molecular descriptors.   The toxicity prediction model according to any one of claims 1 to 6, wherein the toxicity prediction model is a toxicity prediction model constructed by machine learning using normalized molecular descriptor values of a plurality of compounds whose presence or absence of toxicity is known. Prediction method.   The prediction method according to claim 7, wherein the machine learning is at least one machine learning selected from the group consisting of a support vector machine, a Bayesian network, a neural network, Adaboost, random forest, and active learning. Generating a chemical formula of the test compound;
The prediction method according to claim 1, wherein in step (5), the generated chemical formula and the probability are output in association with each other. 2 or more of the test compounds,
The prediction method according to claim 1, wherein in step (5), the probability is output for each test compound.   The prediction method according to any one of claims 1 to 10, wherein in step (5), a determination result of the presence or absence of toxicity of the test compound is output together with the probability.   The prediction method according to any one of claims 1 to 11, wherein the output of step (5) is a display in a tabular format.   The prediction method according to any one of claims 1 to 12, wherein the toxicity is mutagenicity determined by a reverse mutation test using bacteria. An input means for inputting the structural information of the test compound;
A receiving means for receiving structural information of the test compound input by the user;
First generation means for generating a three-dimensional molecular structure having an optimized structure based on the received structure information;
One or more molecular descriptor values including at least one molecular descriptor selected from the group consisting of a three-dimensional molecular descriptor, a four-dimensional molecular descriptor, and a quantum chemical molecular descriptor using the three-dimensional molecular structure. First calculating means for calculating
A calculation means for the toxicity prediction model to calculate the probability of the presence or absence of toxicity of the test compound using the value of the molecular descriptor, and the sum of the probability of toxicity and the probability of non-toxicity is 100%. A second calculation means, and an output means for outputting the calculated probability;
A system for predicting the toxicity of a compound. An input device that functions as the input means;
An arithmetic unit that functions as the first generation unit, the first calculation unit, and the second calculation unit;
An output device functioning as the output means;
A main storage device and a control device for controlling the system;
15. The system of claim 14, comprising:   The system according to claim 15, further comprising an auxiliary storage device in which the program is stored. A process of receiving structural information of the test compound entered by the user;
Based on the received structural information, a process for generating a three-dimensional molecular structure with an optimized structure;
One or more molecular descriptor values including at least one molecular descriptor selected from the group consisting of a three-dimensional molecular descriptor, a four-dimensional molecular descriptor, and a quantum chemical molecular descriptor using the three-dimensional molecular structure. A process of calculating
A process for calculating the probability of the presence or absence of toxicity of the test compound using the value of the molecular descriptor, and a process that is 100% when the probability of toxicity and the probability of non-toxicity are added together And a process for outputting the calculated probability;
A program that causes a computer to execute.   A computer-readable storage medium storing the program according to claim 17.

Images