Retrosynthesis / chemical reaction feasibility prediction methods, systems, media and equipment

Abstract

The present invention provides a method, system, medium and device for predicting the feasibility of retrosynthesis / chemical reaction. The method for predicting the feasibility of chemical reaction includes: obtaining reactants and products of a designed chemical reaction; extracting functional groups of the reactants and reaction templates of the designed chemical reaction; calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants; calculating the selectivity of the designed chemical reaction according to the reaction template; and obtaining a prediction result of the feasibility of the designed chemical reaction according to the compatibility, reactivity and selectivity evaluation. The present invention can realize the feasibility prediction of each chemical reaction in the retrosynthesis path design step, and then can evaluate whether the retrosynthesis chemical reaction automatically generated by the system is feasible.

Description

Technical Field

[0001] The present invention belongs to the technical field of computer chemistry, and relates to a method for predicting the feasibility of a chemical reaction, and in particular to a method, system, medium and equipment for predicting the feasibility of a retrosynthesis / chemical reaction. Background Art

[0002] Retrosynthetic analysis, also known as the disconnection approach, is the most basic and commonly used method for designing organic synthesis routes. Retrosynthetic analysis is a reversible logical thinking method that starts with analyzing the chemical structure of the target molecule. According to the characteristics of the connection mode (chemical bond) between atoms in the molecule, it comprehensively applies the knowledge of organic chemical reaction methods and reaction mechanisms to select appropriate chemical bonds for disconnection, and transform the target molecule into some slightly smaller intermediates; then use these intermediates as new target molecules and disconnect them into smaller intermediates; and so on, until the starting material that can be easily purchased is found. It is a method that starts from the molecular structure of the synthetic product and uses the "disconnection of a chemical bond" analysis method to obtain the required synthetic raw materials (synthons).

[0003] In the process of retrosynthetic analysis, chemists need to spend a lot of time to verify whether each step of the retrosynthetic path is feasible and whether there will be any abnormalities. Even if a feasible retrosynthetic path is obtained, it is impossible to determine whether the path is the best path among various feasible retrosynthetic paths. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, the object of the present invention is to provide a method, system, medium and equipment for predicting the feasibility of retrosynthesis / chemical reaction, which are used to effectively predict the feasibility of chemical reactions in the retrosynthesis path design step.

[0005] To achieve the above-mentioned purpose and other related purposes, the present invention provides a method for predicting the feasibility of a chemical reaction, which comprises: obtaining reactants and products of a designed chemical reaction; extracting functional groups of the reactants and a reaction template of the designed chemical reaction; calculating the compatibility and reactivity of the designed chemical reaction based on the functional groups of the reactants; calculating the selectivity of the designed chemical reaction based on the reaction template; and obtaining a prediction result of the feasibility of the designed chemical reaction based on the compatibility, reactivity and selectivity evaluation.

[0006] In one embodiment of the present invention, an implementation process of extracting the functional groups of the reactants includes: extracting the first adjacent atoms directly bonded to each central atom of the reactants, obtaining the atomic central functional groups of the reactants, and recording the properties of the first adjacent atoms; extracting the two end atoms of each chemical bond of the reactants and the second adjacent atoms directly bonded to the two end atoms, obtaining the chemical bond central functional groups of the reactants, and recording the properties of the second adjacent atoms.

[0007] In one embodiment of the present invention, a process for calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants includes: calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants. i Calculate the designed chemical reaction in the functional group g i Compatibility scores when present are: Wherein, R represents the reaction template of the designed chemical reaction to be predicted, P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database, and P(g i ) represents the functional group g in the reaction database i The compatibility score of the designed chemical reaction calculated based on the set g of each functional group in the reactant is: P(R|g)=ΠP(R|g i ), where g = {g1, g2, … g i ,…} represents the functional group set in the reactants, P(R|g i ) indicates that the designed chemical reaction is in the functional group g i The compatibility score when it exists, П represents the compatibility score for each element g in the functional group set g. i The calculation of the formula on the right side of the symbol is performed and the results are multiplied; according to a functional group not present in the reactant Calculate the designed chemical reactions for functional groups The reactivity score is: Where P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database. The R reaction template is a functional group Compatibility score when present; Represents a functional group in the reaction database Frequency of occurrence; Represents the functional group in the chemical reaction of the R reaction template frequency of occurrence; according to the set g of functional groups near the reaction site in the reactant and the set g of functional groups near the site of the same type of reaction as the reaction template in the reaction template database c The reactivity score of the designed chemical reaction is calculated as: in, represents the set of functional groups in the reaction template database that exist near the reaction site of the chemical reaction belonging to the R reaction template, represents the functional group set g c The functional groups in the reactant do not exist near the reaction site of the reactant; Π represents the functional group set g c Each element The calculation of the formula on the right side of the symbol is performed and the results are multiplied; wherein, the functional group set near the reaction site may refer to, but is not limited to, the set of functional groups to which atoms with a graph theoretical distance of 1 or 2 from the reaction site belong.

[0008] In one embodiment of the present invention, a process for calculating the selectivity of the designed chemical reaction according to the reaction template includes: applying the reaction template to the reactants to find all potential forward reaction sites {s j}; Calculate the feasibility score for each forward reaction site: Wherein, P(R|g) is the compatibility score of the designed chemical reaction calculated based on the set g of functional groups in the reactants, is a set of functional groups near the reaction site in the reactant and a set of functional groups near the reaction site of the chemical reaction belonging to the recommended reaction template R in the reaction template database. c The calculated reactivity score of the designed chemical reaction; the feasibility score of the main site s0 is calculated based on the feasibility scores of all reaction sites: Where F is the reaction feasibility function, F(s0) represents the feasibility score when the reaction occurs at site s0, and F(s0|{s j}) represents the set of other possible reaction sites {s j}, will soon be assembled j}Selectivity is included in the calculation of the feasibility score when the reaction occurs at site s0; the potential positive reaction site {s j}, that is, all the sites in the reactants that may carry out the R reaction template; the feasibility score of the main site s0 is the predicted result of the feasibility of the designed chemical reaction.

[0009] In one embodiment of the present invention, the chemical reaction feasibility prediction method also includes: retrieving similar reactions of the designed chemical reaction; calculating the similarity between the designed chemical reaction and the similar reactions; and obtaining a verification result of the feasibility of the designed chemical reaction based on the prediction result of the feasibility of the designed chemical reaction and the similarity.

[0010] In one embodiment of the present invention, an implementation process of retrieving similar reactions of the designed chemical reaction includes: searching for chemical reactions that are identical or similar to the designed chemical reaction in a chemical reaction database based on search factors; the search factors include reaction template features, reaction site features, functional group features and / or similarity features; the reaction template features include reaction template ID and extended reaction template ID; the reaction site features include main site features and competitive site features; the functional group features include atomic center functional group features and chemical bond center functional group features; the similarity features include but are not limited to information about atoms and their adjacent atoms, the number of atoms, bonding order, bond order, number of rings, adjacency of rings, etc.; the similarity between the searched chemical reaction and the designed chemical reaction is calculated as:

[0011]

[0012] Among them, the two sets of eigenvectors f and g are classified as symmetric (f sym ,g sym ) and asymmetry (f asym ,g asym ); f represents the characteristic vector of the designed chemical reaction, g represents the characteristic vector of the searched chemical reaction; f k Represents the one-dimensional feature data in the multidimensional feature vector f, g k Represents the one-dimensional feature data in the multidimensional feature vector g, that is, the kth search element feature; θ is the preset weight parameter.

[0013] The present invention also provides a method for predicting the feasibility of a retrosynthetic chemical reaction, which comprises: obtaining a retrosynthetic chemical reaction pathway design; the retrosynthetic chemical reaction pathway design comprises at least one retrosynthetic chemical reaction step; obtaining reactants and products of the designed chemical reaction of the retrosynthetic chemical reaction step; extracting the functional groups of the reactants and the reaction template of the designed chemical reaction; calculating the compatibility and reactivity of each functional group of the reactants; calculating the selectivity of the designed chemical reaction according to the reaction template; obtaining a prediction result of the feasibility of the designed chemical reaction according to the compatibility, reactivity and selectivity evaluation; retrieving similar reactions of the designed chemical reaction; calculating the similarity between the designed chemical reaction and the similar reactions; obtaining a verification result of the feasibility of the designed chemical reaction according to the prediction result of the feasibility of the designed chemical reaction and the similarity; obtaining a feasibility prediction result of the retrosynthetic chemical reaction pathway design according to the verification results of each of the retrosynthetic chemical reaction steps.

[0014] The present invention also provides a chemical reaction feasibility prediction system, which includes: a chemical reaction input module, which obtains reactants and products of a designed chemical reaction; a reaction functional group extraction module, which is connected to the chemical reaction input module and extracts the functional groups of the reactants; a reaction template extraction module, which is connected to the chemical reaction input module and extracts the reaction template of the designed chemical reaction; a reactivity calculation module, which is connected to the reaction functional group extraction module and calculates the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants; a selectivity calculation module, which is connected to the reaction template extraction module and calculates the selectivity of the designed chemical reaction according to the reaction template; and a feasibility prediction module, which is respectively connected to the reactivity calculation module and the selectivity calculation module and obtains a prediction result of the feasibility of the designed chemical reaction according to the compatibility, reactivity and selectivity evaluation.

[0015] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the chemical reaction feasibility prediction method or the steps of the retrosynthetic chemical reaction feasibility prediction method are implemented.

[0016] The present invention also provides a device, comprising: a memory storing a computer program; a processor communicating with the memory, for implementing the steps of the method for predicting the feasibility of a retrosynthetic chemical reaction or the steps of the method for predicting the feasibility of a retrosynthetic chemical reaction when calling and executing the computer program.

[0017] As described above, the retrosynthesis / chemical reaction feasibility prediction method, system, medium and device of the present invention have the following beneficial effects:

[0018] The present invention can realize the feasibility prediction of each chemical reaction in the retrosynthetic path design step, and then can evaluate whether the retrosynthetic chemical reaction automatically generated by the system is feasible. In addition, the present invention also realizes efficient similar reaction retrieval, and can consider the similarity of compounds and the fragility of functional groups when searching for similar reactions, so that the feasibility prediction result conforms to the actual situation, improves the reliability of the retrosynthetic analysis of the system, and also provides a basis for selecting a better retrosynthetic path. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1A Shown is a schematic diagram of an implementation flow of the chemical reaction feasibility prediction method described in an embodiment of the present invention.

[0020] Figure 1B Shown is another implementation flow diagram of the chemical reaction feasibility prediction method described in an embodiment of the present invention.

[0021] Figure 2AShown is an exemplary structural schematic diagram of the atomic central functional group described in an embodiment of the present invention.

[0022] Figure 2B Shown is an exemplary structural schematic diagram of the chemical bond center functional group described in an embodiment of the present invention.

[0023] Figure 3 Shown is a schematic diagram of an implementation flow of the method for predicting the feasibility of a retrosynthetic chemical reaction described in an embodiment of the present invention.

[0024] Figure 4 Shown is a schematic diagram of an implementation structure of the chemical reaction feasibility prediction system described in an embodiment of the present invention.

[0025] Figure 5 Shown is another schematic diagram of the implementation structure of the chemical reaction feasibility prediction system described in an embodiment of the present invention.

[0026] Component number description

[0027] 400 Chemical Reaction Feasibility Prediction System

[0028] 410 Chemical Reaction Input Module

[0029] 420 Reaction Functional Group Extraction Module

[0030] 430 Reaction Template Extraction Module

[0031] 440 Reactive Calculation Module

[0032] 450 Selective Computation Module

[0033] 460 Feasibility Prediction Module

[0034] 470 Similar reaction retrieval module

[0035] 480 Similarity calculation module

[0036] 490 Feasibility Verification Module

[0037] Steps S101 to S108

[0038] Steps S301 to S310 DETAILED DESCRIPTION

[0039] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0040] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0041] In the process of retrosynthetic analysis by chemists, the feasibility of each designed reaction needs to be constantly checked. In the process of automatic retrosynthetic path design, for the reaction predicted by the system, it is necessary to be able to predict the feasibility of the reaction, so as to filter out chemical reactions that are impossible to occur and improve the accuracy of the generated route. For the reaction predicted by the system, it is necessary to quickly find similar reactions that can prove the feasibility of this reaction. Based on the above needs, this application is proposed.

[0042] See also Figure 1A As shown, an embodiment of the present invention provides a chemical reaction feasibility prediction method, the chemical reaction feasibility prediction method comprising:

[0043] Step S101, obtaining reactants and products of a designed chemical reaction;

[0044] Step S102, extracting the functional groups of the reactants and the reaction template of the designed chemical reaction;

[0045] Step S103, calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants;

[0046] Step S104, calculating the selectivity of the designed chemical reaction according to the reaction template;

[0047] Step S105, obtaining a prediction result of the feasibility of the designed chemical reaction based on the compatibility, reactivity and selectivity evaluation.

[0048] See also Figure 1B As shown, in one embodiment of the present invention, the chemical reaction feasibility prediction method further includes:

[0049] Step S106, searching for similar reactions to the designed chemical reaction;

[0050] Step S107, calculating the similarity between the designed chemical reaction and the similar reaction;

[0051] Step S108, obtaining a verification result of the feasibility of the designed chemical reaction according to the prediction result of the feasibility of the designed chemical reaction and the similarity.

[0052] The protection scope of the chemical reaction feasibility prediction method described in the present invention is not limited to the execution order of the steps listed in this embodiment. All solutions achieved by adding, reducing or replacing steps in the prior art based on the principles of the present invention are included in the protection scope of the present invention.

[0053] In one embodiment of the present invention, in step S102, an implementation process of extracting the functional group of the reactant includes: extracting the first adjacent atom directly bonded to each central atom of the reactant to obtain the atomic central functional group of the reactant, and recording the properties of the first adjacent atom; extracting the two end atoms of each chemical bond of the reactant and the second adjacent atoms directly bonded to the two end atoms to obtain the chemical bond central functional group of the reactant, and recording the properties of the second adjacent atom.

[0054] Among them, functional groups are atoms or groups of atoms that determine the chemical properties of organic compounds. Common functional groups include olefins, alcohols, phenols, ethers, aldehydes, ketones, etc. Organic chemical reactions mainly occur on functional groups, which play a decisive role in the properties of organic matter. -X, -OH, -CHO, -COOH, -NO2, -SO3H, -NH2, RCO-, these functional groups determine the chemical properties of halogenated hydrocarbons, alcohols or phenols, aldehydes, carboxylic acids, nitro compounds or nitrites, sulfonic acid organic matter, amines, and amides in organic matter.

[0055] Furthermore, in order to determine the feasibility of a chemical reaction, it is necessary to evaluate the compatibility of the functional groups in the reactants with the required reaction conditions. This embodiment uses an automatically extracted functional group template based on two types of graph theory structures: 1) atomic center functional groups; 2) chemical bond center functional groups. Among them, the atomic center functional group is an atom-centered functional group. For each central atom, the adjacent atoms directly bonded to it (i.e., the first adjacent atom) are extracted. All heavy atoms except carbon are extracted, such as Figure 2A The chemical bond center functional group shown is a functional group centered on the chemical bond. For each chemical bond, the atoms at both ends and the adjacent atoms directly bonded to the atoms at both ends (i.e., the second adjacent atoms) are extracted. For all non-single bond chemical bonds, Figure 2B When extracting atoms from functional groups, the following properties of the atoms are recorded: whether they are on an aromatic ring, whether they form a bond with a non-sp3 atom, whether they are located in a ring, etc., to help distinguish functional groups with different stability and compatibility.

[0056] In one embodiment of the present invention, in step S103, a process for calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants includes: calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants. i Calculate the designed chemical reaction in the functional group g i Compatibility scores when present are:

[0057]

[0058] Wherein, R represents the reaction template of the designed chemical reaction to be predicted, P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database, and P(g i ) represents the functional group g in the reaction database i Frequency of occurrence; P(g i |R) represents the functional group g in the chemical reaction of the R reaction template i Frequency of occurrence;

[0059] The compatibility score of the designed chemical reaction calculated based on the set g of each functional group in the reactants is:

[0060] P(R|g)=ΠP(R|g i )

[0061] Where g = {g1, g2, … g i ,…} represents the functional group set in the reactants, P(R|g i ) indicates that the designed chemical reaction is in the functional group g i The compatibility score when it exists, Π represents the compatibility score for each element g in the functional group set g. i The calculation of the formula on the right side of the symbol is performed and the results are multiplied;

[0062] According to a functional group not present in the reactants Calculate the designed chemical reactions for functional groups The reactivity score is:

[0063]

[0064] Where P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database. The R reaction template is a functional group Compatibility score when present; Represents a functional group in the reaction database Frequency of occurrence; Represents the functional group in the chemical reaction of the R reaction template Frequency of occurrence;

[0065] According to the set g of functional groups near the reaction site in the reactant and the set g of functional groups near the site of the same type of reaction as the reaction template in the reaction template database c The reactivity score of the designed chemical reaction is calculated as:

[0066]

[0067] in, represents the set of functional groups in the reaction template database that exist near the reaction site of the chemical reaction belonging to the R reaction template, represents the functional group set g c The functional groups in the reactant do not exist near the reaction site of the reactant; Π represents the functional group set g c Each element The calculation of the formula on the right side of the symbol is performed and the results are multiplied; wherein, the functional group set near the reaction site may refer to, but is not limited to, the set of functional groups to which atoms with a graph theoretical distance of 1 or 2 from the reaction site belong.

[0068] In this application, the symbol i is a general term, indicating the i-th, and has no specific physical meaning. Among them, the R category refers to the category to which the reaction with the same reaction template as the reaction R belongs. For example, if the reaction template of reaction 1 is the same as the reaction template of reaction R, then reaction 1 and reaction R belong to the same category of reactions, and this category can be called R category.

[0069] Furthermore, the feasibility score of the present invention takes into account compatibility and reactivity. The present invention considers the frequency of occurrence of the reactive functional group in the reaction database in the same type of reaction as the predicted reaction when calculating compatibility. If a functional group appears in the predicted reaction but rarely appears in the same type of reaction in the reaction database, it means that there is a problem with the reaction predicted by the system.

[0070] Reaction database is a chemical reaction database, which is generally extracted from chemical reactions with experimental evidence provided by existing literature or / and patent databases, where each reaction entry contains relevant information such as reactant structure, product structure, reaction conditions, operation steps, etc. The chemical reaction database includes reactants, products and chemical reaction data sets that have been verified to be able to undergo chemical reactions.

[0071] The reaction template database consists of reaction templates of various types of chemical reactions and reaction entries in the reaction database of the same type. After extracting the atoms and their bonds that undergo changes in bonding and / or hybridization during the chemical reaction process, the reactions with the same extraction results are clustered into the same type of reactions. The graph structure that reflects the changes in atoms and chemical bonds of this type of reaction is the reaction template.

[0072] In one embodiment of the present invention, in step S104, an implementation process of calculating the selectivity of the designed chemical reaction according to the reaction template includes:

[0073] Apply the reaction template to the reactants to find all potential forward reaction sites {s j};

[0074] Calculate the feasibility score for each forward reaction site:

[0075]

[0076] Wherein, P(R|g) is the compatibility score of the designed chemical reaction calculated based on the set g of functional groups in the reactants, is a set of functional groups near the reaction site in the reactant and a set of functional groups near the reaction site of the chemical reaction belonging to the recommended reaction template R in the reaction template database. c a calculated reactivity score for the designed chemical reaction;

[0077] The feasibility score of the main site s0 is calculated based on the feasibility scores of all reaction sites:

[0078]

[0079] Where F is the reaction feasibility function, F(s0) represents the feasibility score when the reaction occurs at site s0, and F(s0|{s j}) represents the set of other possible reaction sites {s j}, will soon be assembled j}Selectivity is included in the calculation of the feasibility score when the reaction occurs at site s0; the potential positive reaction site {s j} is all the sites in the reactants that may carry out the R reaction template; the feasibility score of the main site s0 is the predicted result of the feasibility of the designed chemical reaction. In this application, the symbol j is a general reference, indicating the jth, and has no specific physical meaning.

[0080] In one embodiment of the present invention, in step S106, an implementation process of retrieving similar reactions of the designed chemical reaction includes: searching for chemical reactions that are identical or similar to the designed chemical reaction in a chemical reaction database based on search factors; the search factors include reaction template features, reaction site features, functional group features or / and similarity features; the reaction template features include reaction template ID and extended reaction template ID; the reaction site features include main site features and competitive site features; the functional group features include atomic center functional group features and chemical bond center functional group features; the similarity features include but are not limited to information on atoms and their adjacent atoms, the number of atoms, bonding order, bond order, number of rings, and / or adjacency of rings.

[0081] In one embodiment of the present invention, in step S107, the similarity between the searched chemical reaction and the designed chemical reaction is calculated as:

[0082]

[0083] Among them, the two sets of eigenvectors f and g are classified as symmetric (f sym ,g sym ) and asymmetry (f asym ,g asym ); f represents the characteristic vector of the designed chemical reaction, g represents the characteristic vector of the searched chemical reaction; f k Represents the one-dimensional feature data in the multidimensional feature vector f, g k Represents the one-dimensional feature data in the multidimensional feature vector g, that is, the kth search factor feature; θ is a preset weight parameter. Among them, the multidimensional feature vector includes but is not limited to feature data of dimensions such as atoms and their adjacent atoms, the number of atoms, bonding order, bond order, number of rings and adjacency relationship of rings; the one-dimensional feature data is any feature data of atoms and their adjacent atoms, the number of atoms, bonding order, bond order, number of rings and adjacency relationship of rings. In this application, the symbol k is a general term, indicating the kth, and has no specific physical meaning.

[0084] There are two cases for the features in the feature vector: one is that a feature in the literature must be exactly the same as that in the designed chemical reaction, that is, it cannot be more or less, and this case is a symmetric feature vector; the other is that features that do not exist in the designed chemical reaction are allowed to appear in the literature, that is, it can be more but not less; the similarity penalty is only calculated when this feature exists in the designed chemical reaction but not in the literature; this case is an asymmetric feature vector.

[0085] After completing feature extraction, the present invention uses an inverted index module to build a database for all reactions in the literature and improve the reaction database, which can greatly improve the efficiency of subsequent retrieval.

[0086] The present invention realizes accurate chemical reaction feasibility assessment, can evaluate whether the reaction output by the retrosynthesis system is feasible, realizes efficient similar reaction retrieval, can consider the similarity of compounds and the fragility of functional groups when searching, and improves the accuracy of the search.

[0087] See also Figure 3 As shown, the embodiment of the present invention also provides a method for predicting the feasibility of a retrosynthetic chemical reaction, which describes the prediction process of the entire retrosynthetic chemical reaction path design, which includes the embodiment of the present invention Figure 1A and Figure 1B The related content; the feasibility prediction method of the retrosynthetic chemical reaction includes:

[0088] S301, obtaining a retrosynthetic chemical reaction pathway design; the retrosynthetic chemical reaction pathway design includes at least one retrosynthetic chemical reaction step.

[0089] S302, obtaining reactants and products of the designed chemical reaction of the retrosynthetic chemical reaction step.

[0090] S303, extracting the functional groups of the reactants and the reaction template of the designed chemical reaction. One implementation process of extracting the functional groups of the reactants includes: extracting the first adjacent atoms directly bonded to each central atom of the reactants, obtaining the atomic central functional groups of the reactants, and recording the properties of the first adjacent atoms; extracting the two end atoms of each chemical bond of the reactants and the second adjacent atoms directly bonded to the two end atoms, obtaining the chemical bond central functional groups of the reactants, and recording the properties of the second adjacent atoms.

[0091] S304, calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants.

[0092] Further, according to a functional group g of the reactant i Calculate the designed chemical reaction in the functional group g i Compatibility scores when present are:

[0093]

[0094] Wherein, R represents the reaction template of the designed chemical reaction to be predicted, P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database, and P(g i ) represents the functional group g in the reaction database i Frequency of occurrence; P(g i |R) represents the functional group g in the chemical reaction of the R reaction template i Frequency of occurrence;

[0095] The compatibility score of the designed chemical reaction calculated based on the set g of each functional group in the reactants is:

[0096] P(R|g)=ΠP(R|g i )

[0097] Where g = {g1, g2, … g i ,…} represents the functional group set in the reactants, P(R|g i ) indicates that the designed chemical reaction is in the functional group g i The compatibility score when it exists, ∏ represents the compatibility score for each element g in the functional group set g. i The calculation of the formula on the right side of the symbol is performed and the results are multiplied;

[0098] According to a functional group not present in the reactants Calculate the designed chemical reactions for functional groups The reactivity score is:

[0099]

[0100] Where P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database. The R reaction template is a functional group Compatibility score when present; Represents a functional group in the reaction database Frequency of occurrence; Represents the functional group in the chemical reaction of the R reaction template Frequency of occurrence;

[0101] According to the set g of functional groups near the reaction site in the reactant and the set g of functional groups near the site of the same type of reaction as the reaction template in the reaction template database c The reactivity score of the designed chemical reaction is calculated as:

[0102]

[0103] in, represents the set of functional groups in the reaction template database that exist near the reaction site of the chemical reaction belonging to the R reaction template, represents the functional group set g c The functional groups in the reactant do not exist near the reaction site of the reactant; П represents the functional group set g c Each element The calculation of the formula on the right side of the symbol is performed and the results are multiplied; wherein, the functional group set near the reaction site may refer to, but is not limited to, the set of functional groups to which atoms with a graph theoretical distance of 1 or 2 from the reaction site belong.

[0104] S305, calculating the selectivity of the designed chemical reaction according to the reaction template. Wherein, one implementation process of calculating the selectivity of the designed chemical reaction according to the reaction template includes:

[0105] Apply the reaction template to the reactants to find all potential forward reaction sites {s j};

[0106] Calculate the feasibility score for each forward reaction site:

[0107]

[0108] Wherein, P(R|g) is the compatibility score of the designed chemical reaction calculated based on the set g of functional groups in the reactants, is a set of functional groups near the reaction site in the reactant and a set of functional groups near the reaction site of the chemical reaction belonging to the recommended reaction template R in the reaction template database. c a calculated reactivity score for the designed chemical reaction;

[0109] The feasibility score of the main site s0 is calculated based on the feasibility scores of all reaction sites:

[0110]

[0111] Where F is the reaction feasibility function, F(s0) represents the feasibility score when the reaction occurs at site s0, and F(s0|{s j}) represents the set of other possible reaction sites {s j}, will soon be assembled j}Selectivity is included in the calculation of the feasibility score when the reaction occurs at site s0; the potential positive reaction site {s j} is all the sites in the reactants that may carry out the R reaction template; the feasibility score of the main site s0 is the predicted result of the feasibility of the designed chemical reaction. In this application, the symbol j is a general reference, indicating the jth, and has no specific physical meaning.

[0112] S306, obtaining a prediction result of the feasibility of the designed chemical reaction based on the compatibility, reactivity and selectivity evaluation.

[0113] S307, searching for similar reactions of the designed chemical reaction. Further, searching for chemical reactions that are identical or similar to the designed chemical reaction based on search factors; the search factors include reaction template features, reaction site features, functional group features, and / or similarity features; the reaction template features include reaction template ID and extended reaction template ID; the reaction site features include main site features and competitive site features; the functional group features include atomic center functional group features and chemical bond center functional group features; the similarity features include but are not limited to information of atoms and their adjacent atoms, the number of atoms, bonding order, bond order, number of rings, and / or adjacency of rings.

[0114] S308, calculating the similarity between the designed chemical reaction and the similar reaction. The similarity between the searched chemical reaction and the designed chemical reaction is calculated as:

[0115]

[0116] Among them, the two sets of eigenvectors f and g are classified as symmetric (f sym ,g sym ) and asymmetry (f asym ,g asym ); f represents the characteristic vector of the designed chemical reaction, g represents the characteristic vector of the searched chemical reaction; f k Represents the one-dimensional feature data in the multidimensional feature vector f, g k Represents the one-dimensional feature data in the multidimensional feature vector g, that is, the kth search element feature; θ is the preset weight parameter.

[0117] S309: Obtain a verification result of the feasibility of the designed chemical reaction according to the prediction result of the feasibility of the designed chemical reaction and the similarity.

[0118] S310, obtaining a feasibility prediction result of the retrosynthetic chemical reaction pathway design according to the verification results of each of the retrosynthetic chemical reaction steps.

[0119] The protection scope of the feasibility prediction method for retrosynthetic chemical reactions described in the present invention is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present invention are included in the protection scope of the present invention.

[0120] The present invention also provides a chemical reaction feasibility prediction system, which can implement the chemical reaction feasibility prediction method described in the present invention. However, the implementation device of the chemical reaction feasibility prediction method described in the present invention includes but is not limited to the structure of the chemical reaction feasibility prediction system listed in this embodiment. All structural deformations and replacements of the prior art made according to the principles of the present invention are included in the protection scope of the present invention.

[0121] See also Figure 4 As shown, an embodiment of the present invention also provides a chemical reaction feasibility prediction system, and the chemical reaction feasibility prediction system 400 includes: a chemical reaction input module 410, a reaction functional group extraction module 420, a reaction template extraction module 430, a reactivity calculation module 440, a selectivity calculation module 450, and a feasibility prediction module 460.

[0122] The chemical reaction input module 410 obtains reactants and products of a designed chemical reaction.

[0123] The reaction functional group extraction module 420 is connected to the chemical reaction input module 410 to extract the functional groups of the reactants. The reaction functional group extraction module 420 extracts the first adjacent atom directly bonded to each central atom of the reactants to obtain the atomic central functional group of the reactants and records the properties of the first adjacent atom; extracts the two end atoms of each chemical bond of the reactants and the second adjacent atoms directly bonded to the two end atoms to obtain the chemical bond central functional group of the reactants and records the properties of the second adjacent atom.

[0124] The reaction template extraction module 430 is connected to the chemical reaction input module 410 to extract the reaction template of the designed chemical reaction.

[0125] The reactivity calculation module 440 is connected to the reaction functional group extraction module 420 to calculate the compatibility and reactivity of each functional group of the reactant. i Calculate the designed chemical reaction in the functional group g i Compatibility scores when present are:

[0126]

[0127] Wherein, R represents the reaction template of the designed chemical reaction to be predicted, P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database, and P(g i ) represents the functional group g in the reaction database i Frequency of occurrence; P(g i |R) represents the functional group g in the chemical reaction of the R reaction templatei Frequency of occurrence.

[0128] The compatibility score of the designed chemical reaction calculated by the reactivity calculation module 440 according to the set g of each functional group in the reactant is:

[0129] P(R|g)=ΠP(R|g i )

[0130] Where g = {g1, g2, … g i ,…} represents the functional group set in the reactants, P(R|g i ) indicates that the designed chemical reaction is in the functional group g i The compatibility score when it exists, ∏ represents the compatibility score for each element g in the functional group set g. i The calculation of the formula to the right of the symbol is performed and the results are multiplied.

[0131] The reactivity calculation module 440 calculates the reactivity of a functional group that is not present in the reactant. Calculate the designed chemical reactions for functional groups The reactivity score is:

[0132]

[0133] Where P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database. The R reaction template is a functional group Compatibility score when present; Represents a functional group in the reaction database Frequency of occurrence; Represents the functional group in the chemical reaction of the R reaction template Frequency of occurrence.

[0134] The reactivity calculation module 440 calculates the reaction rate of the reaction template based on the set g of functional groups near the reaction site in the reactant and the set g of functional groups near the site of the same type of reaction as the reaction template in the reaction template database. c The reactivity score of the designed chemical reaction is calculated as:

[0135]

[0136] in, represents the set of functional groups in the reaction template database that exist near the reaction site of the chemical reaction belonging to the R reaction template, represents the functional group set g c The functional groups in the reactant do not exist near the reaction site of the reactant; ∏ represents the functional group set g c Each element The calculation of the formula on the right side of the symbol is performed and the results are multiplied; wherein, the functional group set near the reaction site may refer to, but is not limited to, the set of functional groups to which atoms with a graph theoretical distance of 1 or 2 from the reaction site belong.

[0137] The selectivity calculation module 450 is connected to the reaction template extraction module 430 and calculates the selectivity of the designed chemical reaction according to the reaction template. The selectivity calculation module 450 applies the reaction template to the reactants to find all potential positive reaction sites {s j}; Calculate the feasibility score for each forward reaction site: Wherein, P(R|g) is the compatibility score of the designed chemical reaction calculated based on the set g of functional groups in the reactants, is a set of functional groups near the reaction site in the reactant and a set of functional groups near the reaction site of the chemical reaction belonging to the recommended reaction template R in the reaction template database. c The calculated reactivity score of the designed chemical reaction; the feasibility score of the main site s0 is calculated based on the feasibility scores of all reaction sites: Where F is the reaction feasibility function, F(s0) represents the feasibility score when the reaction occurs at site s0, and F(s0|{s j}) represents the set of other possible reaction sites {s j}, will soon be assembled j}Selectivity is included in the calculation of the feasibility score when the reaction occurs at site s0; the potential positive reaction site {s j} is all the sites in the reactants that may carry out the R reaction template; the feasibility score of the main site s0 is the predicted result of the feasibility of the designed chemical reaction. In this application, the symbol j is a general reference, indicating the jth, and has no specific physical meaning.

[0138] Apply the reaction template to the reactants to find all potential forward reaction sites {s j}; Calculate the feasibility score F(s) of each reaction site j ); The feasibility score of the main site s0 is calculated based on the feasibility scores of all reaction sites: Where F is the reaction feasibility function, F(s0) represents the feasibility score when the reaction occurs at site s0, and F(s0|{s j}) represents the set of other possible reaction sites {s j}, that is, the feasibility score of the reaction occurring at site s0 after the selectivity is taken into account; the potential positive reaction site {s j} is all the sites in the reactants that may carry out the reaction template, and the main site s0 is the site for calculating the final feasibility score taking into account the selectivity. The feasibility score of the main site can reflect the prediction result of the feasibility of the designed chemical reaction.

[0139] The feasibility prediction module 460 is connected to the reactivity calculation module 440 and the selectivity calculation module 450 respectively, and obtains the prediction result of the feasibility of the designed chemical reaction according to the compatibility and selectivity evaluation.

[0140] See also Figure 5 As shown, the chemical reaction feasibility prediction system 400 further includes: a similar reaction retrieval module 470 , a similarity calculation module 480 , and a feasibility verification module 490 .

[0141] The similar reaction retrieval module 470 is connected to the reaction template extraction module 430 for communication, and retrieves similar reactions of the designed chemical reaction based on the reaction database. Further, chemical reactions identical or similar to the designed chemical reaction are searched based on search factors; the search factors include reaction template features, reaction site features, functional group features, or / and similarity features; the reaction template features include reaction template ID and extended reaction template ID; the reaction site features include main site features and competitive site features; the functional group features include atomic center functional group features and chemical bond center functional group features; the similarity features include but are not limited to information of atoms and their adjacent atoms, number of atoms, bonding order, bond order, number of rings, or / and adjacency of rings.

[0142] The similarity calculation module 480 is connected to the similar reaction search module 470 for calculating the similarity between the designed chemical reaction and the similar reaction. The similarity between the searched chemical reaction and the designed chemical reaction is calculated as:

[0143]

[0144] Among them, the two sets of eigenvectors f and g are classified as symmetric (f sym ,g sym ) and asymmetry (f asym ,g asym ); f represents the characteristic vector of the designed chemical reaction, g represents the characteristic vector of the searched chemical reaction; f k Represents the one-dimensional feature data in the multidimensional feature vector f, g k Represents the one-dimensional feature data in the multidimensional feature vector g, that is, the kth search element feature; θ is the preset weight parameter.

[0145] The feasibility verification module 490 is in communication with the feasibility prediction module 460 and the similarity calculation module 480 respectively, and obtains the verification result of the feasibility of the designed chemical reaction according to the prediction result of the feasibility of the designed chemical reaction and the similarity.

[0146] It should be noted that it should be understood that the division of the various modules of the above system is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software called by processing elements, or they can all be implemented in the form of hardware, or some modules can be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example: the x module can be a separately established processing element, or it can be integrated in a certain chip of the above system. In addition, the x module can also be stored in the memory of the above system in the form of program code, and called and executed by a certain processing element of the above system. The implementation of other modules is similar. These modules can be integrated together in whole or in part, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in the form of software.

[0147] The present invention also provides a device, comprising: a memory storing a computer program; a processor communicating with the memory, for implementing the steps of the method for predicting the feasibility of a retrosynthetic chemical reaction or the steps of the method for predicting the feasibility of a retrosynthetic chemical reaction when calling and executing the computer program.

[0148] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for predicting the feasibility of a chemical reaction or the method for predicting the feasibility of a retrosynthetic chemical reaction described in the present invention.

[0149] The computer programs described in this application can be downloaded from a computer-readable storage medium to various computing / processing devices, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network.

[0150] In summary, the present invention effectively overcomes various shortcomings of the prior art and has high industrial utilization value.

[0151] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for predicting the feasibility of a chemical reaction, characterized in that: The chemical reaction feasibility prediction method comprises: Obtain the reactants and products of designed chemical reactions; extracting the functional groups of the reactants and the reaction template of the designed chemical reaction; Calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants; calculating the selectivity of the designed chemical reaction according to the reaction template; Obtaining a prediction result of the feasibility of the designed chemical reaction based on the compatibility, reactivity and selectivity evaluation; A process for calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants includes: According to the functional group g of the reactant i The compatibility score of the designed chemical reaction in the presence of functional group gi is calculated as: Wherein, R represents the reaction template of the designed chemical reaction to be predicted, P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database, and P(g i ) represents the functional group g in the reaction database i Frequency of occurrence; P(g i |R) represents the functional group g in the chemical reaction of the R reaction template i Frequency of occurrence; The compatibility score of the designed chemical reaction calculated based on the set g of each functional group in the reactants is: P(R|g)=ΠP(R|g i ) Where g = {g1, g2, ... g i , ...} represents the functional group set in the reactants, P(R|g i ) indicates that the designed chemical reaction is in the functional group g i The compatibility score when it exists, Π represents the compatibility score for each element g in the functional group set g. i The calculation of the formula on the right side of the symbol is performed and the results are multiplied; According to a functional group not present in the reactants Calculate the designed chemical reactions for functional groups The reactivity score is: Where P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database. The R reaction template is a functional group Compatibility score when present; Represents a functional group in the reaction database Frequency of occurrence; Represents the functional group in the chemical reaction of the R reaction template Frequency of occurrence; According to the set g of functional groups near the reaction site in the reactant and the set g of functional groups near the site of the same type of reaction as the reaction template in the reaction template database c The reactivity score of the designed chemical reaction is calculated as: in, represents the set of functional groups in the reaction template database that exist near the reaction site of the chemical reaction belonging to the R reaction template, represents the functional group set g c The functional groups in the reactant do not exist near the reaction site of the reactant; Π represents the functional group set g c Each element The calculation of the formula on the right side of the symbol is performed and the results are multiplied; wherein, the functional group set near the reaction site may refer to, but is not limited to, the set of functional groups to which atoms with a graph theoretical distance of 1 or 2 from the reaction site belong.

2. The method for predicting the feasibility of a chemical reaction according to claim 1, characterized in that: A process for extracting the functional groups of the reactants includes: Extracting the first neighboring atom directly bonded to each central atom of the reactant, obtaining the atomic central functional group of the reactant, and recording the properties of the first neighboring atom; The two end atoms of each chemical bond of the reactant and the second adjacent atom directly bonded to the two end atoms are extracted to obtain the central functional group of the chemical bond of the reactant, and the properties of the second adjacent atom are recorded.

3. The method for predicting the feasibility of a chemical reaction according to claim 1, characterized in that: An implementation process of calculating the selectivity of the designed chemical reaction according to the reaction template includes: Apply the reaction template to the reactants to find all potential forward reaction sites {s j }; Calculate the feasibility score for each forward reaction site: Wherein, P(R|g) is the compatibility score of the designed chemical reaction calculated based on the set g of functional groups in the reactants, is a set of functional groups near the reaction site in the reactant and a set of functional groups near the reaction site of the chemical reaction belonging to the recommended reaction template R in the reaction template database. c a calculated reactivity score for the designed chemical reaction; The feasibility score of the main site s0 is calculated based on the feasibility scores of all reaction sites: Where F is the reaction feasibility function, F(s0) represents the feasibility score when the reaction occurs at site s0, and F(s0|{s j }) represents the set of other possible reaction sites {s j }, will soon be assembled j }Selectivity is included in the calculation of the feasibility score when the reaction occurs at site s0; the potential positive reaction site {s j }, that is, all the sites in the reactants that may carry out the R reaction template; the feasibility score of the main site s0 is the predicted result of the feasibility of the designed chemical reaction.

4. The method for predicting the feasibility of a chemical reaction according to claim 1, characterized in that: The chemical reaction feasibility prediction method also includes: Retrieving similar reactions of the designed chemical reaction; Calculating the similarity between the designed chemical reaction and the similar reaction; A verification result of the feasibility of the designed chemical reaction is obtained according to the prediction result of the feasibility of the designed chemical reaction and the similarity.

5. The method for predicting the feasibility of a chemical reaction according to claim 4, characterized in that: An implementation process of retrieving similar reactions of the designed chemical reaction includes: In a chemical reaction database, a chemical reaction identical or similar to the designed chemical reaction is searched based on search factors; the search factors include reaction template features, reaction site features, functional group features or / and similarity features; the reaction template features include reaction template ID and extended reaction template ID; the reaction site features include main site features and competitive site features; the functional group features include atomic center functional group features and chemical bond center functional group features; the similarity features include but are not limited to information of atoms and their adjacent atoms, number of atoms, bonding order, bond order, number of rings, and / or adjacency relationship of rings; The similarity between the searched chemical reaction and the designed chemical reaction is calculated as: Among them, the two sets of eigenvectors f and g are classified as symmetric (f sym , g sym ) and asymmetry (f asym , g asym ); f represents the characteristic vector of the designed chemical reaction, g represents the characteristic vector of the searched chemical reaction; f k Represents the one-dimensional feature data in the multidimensional feature vector f, g k Represents the one-dimensional feature data in the multidimensional feature vector g, that is, the kth search element feature; θ is the preset weight parameter.

6. A method for predicting the feasibility of a retrosynthetic chemical reaction, characterized in that: The feasibility prediction method for the retrosynthetic chemical reaction comprises: Obtaining a retrosynthetic chemical reaction pathway design; the retrosynthetic chemical reaction pathway design includes at least one retrosynthetic chemical reaction step; Obtaining reactants and products of the designed chemical reaction of the retrosynthetic chemical reaction step; extracting the functional groups of the reactants and the reaction template of the designed chemical reaction; Calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants; calculating the selectivity of the designed chemical reaction according to the reaction template; Obtaining a prediction result of the feasibility of the designed chemical reaction based on the compatibility, reactivity and selectivity evaluation; Retrieving similar reactions of the designed chemical reaction; Calculating the similarity between the designed chemical reaction and the similar reaction; Obtaining a verification result of the feasibility of the designed chemical reaction according to the prediction result of the feasibility of the designed chemical reaction and the similarity; Obtaining a feasibility prediction result of the retrosynthetic chemical reaction pathway design according to the verification results of each of the retrosynthetic chemical reaction steps; A process for calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants includes: According to the functional group g of the reactant i The compatibility score of the designed chemical reaction in the presence of functional group gi is calculated as: Wherein, R represents the reaction template of the designed chemical reaction to be predicted, P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database, and P(g i ) represents the functional group g in the reaction database i Frequency of occurrence; P(g i |R) represents the functional group g in the chemical reaction of the R reaction template i Frequency of occurrence; The compatibility score of the designed chemical reaction calculated based on the set g of each functional group in the reactants is: P(R|g)=ΠP(R|g i ) Where g = {g1, g2, ... g i , ...} represents the functional group set in the reactants, P(R|g i ) indicates that the designed chemical reaction is in the functional group g i The compatibility score when it exists, Π represents the compatibility score for each element g in the functional group set g. i The calculation of the formula on the right side of the symbol is performed and the results are multiplied; According to a functional group not present in the reactants Calculate the designed chemical reactions for functional groups The reactivity score is: Where P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database. The R reaction template is a functional group Compatibility score when present; Represents a functional group in the reaction database Frequency of occurrence; Represents the functional group in the chemical reaction of the R reaction template Frequency of occurrence; According to the set g of functional groups near the reaction site in the reactant and the set g of functional groups near the site of the same type of reaction as the reaction template in the reaction template database c The reactivity score of the designed chemical reaction is calculated as: in, represents the set of functional groups in the reaction template database that exist near the reaction site of the chemical reaction belonging to the R reaction template, represents the functional group set g c The functional groups in the reactant do not exist near the reaction site of the reactant; Π represents the functional group set g c Every element The calculation of the formula on the right side of the symbol is performed and the results are multiplied; wherein, the functional group set near the reaction site may refer to, but is not limited to, the set of functional groups to which atoms with a graph theoretical distance of 1 or 2 from the reaction site belong.

7. A chemical reaction feasibility prediction system, characterized in that: The chemical reaction feasibility prediction system comprises: Chemical reaction input module, which obtains the reactants and products of the designed chemical reaction; A reaction functional group extraction module, connected to the chemical reaction input module, extracting the functional groups of the reactants; A reaction template extraction module, connected to the chemical reaction input module, extracts the reaction template of the designed chemical reaction; A reactivity calculation module is connected to the reaction functional group extraction module, and calculates the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants; an implementation process of calculating the compatibility and reactivity of the designed chemical reaction according to the functional groups of the reactants includes: According to the functional group g of the reactant i Calculate the designed chemical reaction in the functional group g i Compatibility scores when present are: Wherein, R represents the reaction template of the designed chemical reaction to be predicted, P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database, and P(g i ) represents the functional group g in the reaction database i Frequency of occurrence; P(g i |R) represents the functional group g in the chemical reaction of the R reaction template i Frequency of occurrence; The compatibility score of the designed chemical reaction calculated based on the set g of each functional group in the reactants is: P(R|g)=ΠP(R|g i ) Where g = {g1, g2, ... g i , ...} represents the functional group set in the reactants, P(R|g i ) indicates that the designed chemical reaction is in the functional group g i The compatibility score when it exists, Π represents the compatibility score for each element g in the functional group set g. i The calculation of the formula on the right side of the symbol is performed and the results are multiplied; According to a functional group not present in the reactants Calculate the designed chemical reactions for functional groups The reactivity score is: Where P(R) represents the frequency of occurrence of chemical reactions belonging to the R reaction template in the reaction database. The R reaction template is a functional group Compatibility score when present; Represents a functional group in the reaction database Frequency of occurrence; Represents the functional group in the chemical reaction of the R reaction template Frequency of occurrence; According to the set g of functional groups near the reaction site in the reactant and the set g of functional groups near the site of the same type of reaction as the reaction template in the reaction template database c The reactivity score of the designed chemical reaction is calculated as: in, represents the set of functional groups in the reaction template database that exist near the reaction site of the chemical reaction belonging to the R reaction template, represents the functional group set g c The functional groups in the reactant do not exist near the reaction site of the reactant; Π represents the functional group set g c Each element The formula on the right side of the symbol is calculated and the results are multiplied; wherein the functional group set near the reaction site may refer to, but is not limited to, a set of functional groups to which atoms with a graph theoretical distance of 1 and 2 from the reaction site belong; a selectivity calculation module, connected to the reaction template extraction module, calculates the selectivity of the designed chemical reaction according to the reaction template; A feasibility prediction module is connected to the reactivity calculation module and the selectivity calculation module respectively, and obtains a prediction result of the feasibility of the designed chemical reaction according to the compatibility, reactivity and selectivity evaluation.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 5 or the steps of the method according to claim 6 are implemented.

9. A device, characterized in that: The device comprises: A memory storing a computer program; A processor is communicatively connected to the memory, and implements the steps of the method according to any one of claims 1 to 5 or the steps of the method according to claim 6 when calling and executing the computer program.

Images