CN112562869A

CN112562869A - Drug combination safety evaluation system, method and device

Info

Publication number: CN112562869A
Application number: CN202110205796.4A
Authority: CN
Inventors: 商洪才; 陈昭; 田贵华; 张晓雨; 郑蕊; 潘海娥; 邱瑞瑾; 蒋寅; 魏煦旭; 万思琦; 刘文静
Original assignee: DONGZHIMEN HOSPITAL AFFILIATED TO BEIJING UNIVERSITY OF CHINESE MEDICINE
Current assignee: DONGZHIMEN HOSPITAL AFFILIATED TO BEIJING UNIVERSITY OF CHINESE MEDICINE
Priority date: 2021-02-24
Filing date: 2021-02-24
Publication date: 2021-03-26

Abstract

The application discloses a drug combination safety evaluation system, a method and a device, wherein the system comprises an input module, a theme database and a safety evaluation module; the input module is used for obtaining a first medicament and a second medicament to be combined; the theme database is used for determining each chemical component in the first medicament and each chemical component in the second medicament according to the first medicament and the second medicament; the safety evaluation module is used for obtaining a multi-organ damage prediction result of each chemical component in the first medicine, a multi-organ damage prediction result of each chemical component in the second medicine, each predicted metabolite of each chemical component in the first medicine and each chemical component in the second medicine in an interaction manner, and a multi-organ damage prediction result of each predicted metabolite according to each chemical component in the first medicine, each chemical component in the second medicine and the medicine combination multi-organ damage prediction model, and the multi-organ damage prediction results are used as safety evaluation results of the first medicine and the second medicine to be combined.

Description

Drug combination safety evaluation system, method and device

Technical Field

The application relates to the technical field of data processing and analysis, in particular to a drug combination safety evaluation system, method and device.

Background

The combination of drugs has a long history during the course of drug therapy. The drug combination is that two or more than two drugs are applied simultaneously or sequentially for achieving the purpose of treatment, mainly for increasing the curative effect of the drugs or for relieving the toxic and side effect of the drugs. However, the drug combination may also produce unsafe results such as multiple organ injuries, and therefore, it is necessary to determine whether the drug combination is safe, so as to provide a data base for realizing the drug combination.

Currently, users such as clinicians and medical researchers can know the safety information of drug combination through published literature, network information, medical books, and the like. However, the inventor has found through research that the published literature, network information, medical books and the like all record clinically known drug combination safety information, namely, the results of damage to each chemical component and corresponding multiple organs in the combined drugs; based on this, the method cannot know the unknown drug combination safety information, that is, whether the drug which is not used in combination causes multiple organ damage.

Disclosure of Invention

In view of this, embodiments of the present application provide a system, a method, and a device for drug combination safety evaluation, which enable prediction of whether an unconjugated drug causes multiple organ damage, thereby determining unknown drug combination safety information, and providing a data basis for unknown drug combination to reduce an unknown drug combination risk.

In a first aspect, an embodiment of the present application provides a drug combination safety evaluation system, where the system includes: the system comprises an input module, a theme database and a safety evaluation module;

the input module is used for obtaining a first medicament and a second medicament to be combined;

the theme database is used for determining each chemical component in the first medicament and each chemical component in the second medicament according to the first medicament and the second medicament;

the safety evaluation module is used for obtaining a multi-organ damage prediction result of each chemical component in the first medicine and a multi-organ damage prediction result of each chemical component in the second medicine according to each chemical component in the first medicine, each chemical component in the second medicine and a medicine combination multi-organ damage prediction model, and taking the multi-organ damage prediction results as safety evaluation results of the first medicine and the second medicine to be combined;

the drug combination multi-organ injury prediction model is obtained by pre-training a neural network according to each chemical component in the drug and the corresponding multi-organ injury result.

Optionally, the first medicine is a traditional Chinese medicine, the second medicine is a western medicine, and the theme database comprises a traditional Chinese medicine basic database, a traditional Chinese medicine compound basic database, a traditional Chinese medicine chemical composition database and a western medicine chemical composition database.

Optionally, the theme database is specifically configured to query the traditional Chinese medicine basic database or the traditional Chinese medicine compound basic database according to the traditional Chinese medicine, determine each chemical component in the traditional Chinese medicine, query the traditional Chinese medicine chemical component database according to each chemical component in the traditional Chinese medicine, and determine a simplified molecular linear input specification SMILES code of each chemical component in the traditional Chinese medicine; querying the chemical component database of the western medicine according to the western medicine to determine SMILES (small entry loss) codes of all chemical components in the western medicine;

correspondingly, the safety evaluation module is specifically configured to obtain feature vectors of the chemical components in the traditional Chinese medicine and feature vectors of the chemical components in the western medicine according to the SMILES codes of the chemical components in the traditional Chinese medicine, the SMILES codes of the chemical components in the western medicine and the embedded layer in the medicine combined multi-organ injury prediction model; and obtaining a multi-organ damage prediction result of each chemical component in the traditional Chinese medicine and a multi-organ damage prediction result of each chemical component in the western medicine according to the feature vectors of each chemical component in the traditional Chinese medicine, the feature vectors of each chemical component in the western medicine and a prediction network in the medicine combined multi-organ damage prediction model, and taking the multi-organ damage prediction results as safety evaluation results of the traditional Chinese medicine and the western medicine to be combined.

Optionally, the safety evaluation module is further configured to obtain each predicted metabolite of interaction between each chemical component in the traditional Chinese medicine and each chemical component in the western medicine according to each chemical component in the traditional Chinese medicine, each chemical component in the western medicine, a preset metabolic enzyme and the medicine combined multi-organ injury prediction model; and obtaining a multi-organ injury prediction result of each predicted metabolite according to the multi-organ injury prediction model combining each predicted metabolite and the medicine, and taking the multi-organ injury prediction result as a safety evaluation result of the traditional Chinese medicine and the western medicine to be combined.

Optionally, the safety evaluation module is further specifically configured to obtain the SMILES code of each predicted metabolite by using a preset chemical editor, and obtain the feature vector of each predicted metabolite according to the SMILES code of each predicted metabolite and the embedded layer in the drug combination multi-organ injury prediction model; and obtaining the multi-organ injury prediction result of each predicted metabolite according to the feature vector of each predicted metabolite and a prediction network in the drug combined multi-organ injury prediction model.

Optionally, the subject database further comprises a traditional Chinese medicine and western medicine combined clinical database, a medicine interaction database in combination of traditional Chinese medicine and western medicine, a traditional Chinese medicine and western medicine combined metabolic reaction database, a traditional Chinese medicine and western medicine combined disease database, a multi-organ drug injury database and a multi-organ injury related pathway database; correspondingly, the system further comprises a search module, the search module at least comprises a combined Chinese and Western medicine search submodule, and the combined Chinese and Western medicine search submodule is used for searching the clinical database of combined Chinese and Western medicine, the drug interaction database of combined Chinese and Western medicine, the metabolic reaction database of combined Chinese and Western medicine, the disease database of combined Chinese and Western medicine, the multi-organ drug injury database and the multi-organ injury related pathway database according to the traditional Chinese medicine and the western medicine.

Optionally, the search module further includes a traditional Chinese medicine search submodule, a chemical structural formula search submodule, and/or an advanced search submodule.

Optionally, the system further includes a visualization module, and the visualization module is configured to visually display a safety evaluation result of the first drug and the second drug to be used in combination.

In a second aspect, an embodiment of the present application provides a method for drug combination safety evaluation, where the method includes:

obtaining a first drug and a second drug to be combined;

determining each chemical component in the first medicament and each chemical component in the second medicament according to the first medicament and the second medicament;

obtaining a multi-organ injury prediction result of each chemical component in the first medicament and a multi-organ injury prediction result of each chemical component in the second medicament according to each chemical component in the first medicament, each chemical component in the second medicament and a medicament combination multi-organ injury prediction model, and using the multi-organ injury prediction results as safety evaluation results of the first medicament and the second medicament to be combined;

In a third aspect, an embodiment of the present application provides a drug combination safety evaluation device, where the device includes:

a first obtaining unit for obtaining a first drug and a second drug to be combined;

a determination unit configured to determine each chemical component in the first medicine and each chemical component in the second medicine according to the first medicine and the second medicine;

a second obtaining unit, configured to obtain a multi-organ injury prediction result of each chemical component in the first drug and a multi-organ injury prediction result of each chemical component in the second drug according to each chemical component in the first drug, each chemical component in the second drug, and a drug combination multi-organ injury prediction model, where the multi-organ injury prediction results are used as safety evaluation results of the first drug and the second drug to be combined;

Compared with the prior art, the method has the advantages that:

by adopting the technical scheme of the embodiment of the application, the drug combination safety evaluation system comprises an input module, a theme database and a safety evaluation module; the input module is used for obtaining a first medicament and a second medicament to be combined; the theme database is used for determining each chemical component in the first medicament and each chemical component in the second medicament according to the first medicament and the second medicament; the safety evaluation module is used for obtaining a multi-organ damage prediction result of each chemical component in the first medicine and a multi-organ damage prediction result of each chemical component in the second medicine according to each chemical component in the first medicine, each chemical component in the second medicine and the medicine combination multi-organ damage prediction model, and using the multi-organ damage prediction results as safety evaluation results of the first medicine and the second medicine to be combined; the drug combination multi-organ injury prediction model is obtained by pre-training a neural network according to each chemical component in the drug and a corresponding multi-organ injury result.

Therefore, the drug combination safety evaluation system firstly determines each chemical component in the drugs to be combined through the theme database, then obtains the multi-organ injury prediction result of each chemical component through the drug combination multi-organ injury prediction model for fully learning the incidence relation between each chemical component in the drugs and the corresponding multi-organ injury result in the safety evaluation module, and realizes prediction on whether the chemical components in the unknown drug combination cause multi-organ injury, so that the unknown drug combination safety information is determined, a data base is provided for unknown drug combination, and the unknown drug combination risk is reduced.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments of the present application will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic diagram of a system framework related to an application scenario in an embodiment of the present application;

FIG. 2 is a schematic structural diagram of a drug combination safety evaluation system according to an embodiment of the present disclosure;

FIG. 3 is a schematic structural diagram of another drug combination safety evaluation system provided in the embodiments of the present application;

FIG. 4 is a schematic structural diagram of another drug combination safety evaluation system provided in the embodiments of the present application;

FIG. 5 is a schematic flow chart of a method for drug combination safety evaluation provided in an embodiment of the present application;

fig. 6 is a schematic structural diagram of a device for drug combination safety evaluation provided in an embodiment of the present application.

Detailed Description

In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art without any inventive work according to the embodiments of the present application are within the scope of the present application.

With the increasing application of drug combination in the drug treatment process, drug combination becomes the focus of research attention of users such as clinicians and medical researchers. At present, the safety information of drug combination can be known through published literature, network information, medical books and the like. However, the inventor has found through research that the published literature, network information, medical books and the like all record chemical components and corresponding multiple organ injury results in medicines which are clinically combined; based on this, it is unknown whether the unknown drug combination causes multiple organ damage.

In order to solve the problem, in the embodiment of the application, the drug combination safety evaluation system comprises an input module, a theme database and a safety evaluation module; the input module is used for obtaining a first medicament and a second medicament to be combined; the theme database is used for determining each chemical component in the first medicament and each chemical component in the second medicament according to the first medicament and the second medicament; the safety evaluation module is used for obtaining a multi-organ damage prediction result of each chemical component in the first medicine and a multi-organ damage prediction result of each chemical component in the second medicine according to each chemical component in the first medicine, each chemical component in the second medicine and the medicine combination multi-organ damage prediction model, and using the multi-organ damage prediction results as safety evaluation results of the first medicine and the second medicine to be combined; the drug combination multi-organ injury prediction model is obtained by pre-training a neural network according to each chemical component in the drug and a corresponding multi-organ injury result.

The drug combination safety evaluation system firstly determines each chemical component in the drugs to be combined through the theme database, then obtains the multi-organ damage prediction result of each chemical component through the drug combination multi-organ damage prediction model for fully learning the incidence relation between each chemical component in the drugs and the corresponding multi-organ damage result in the safety evaluation module, and realizes prediction on whether the chemical components in the unknown drug combination cause multi-organ damage or not, thereby determining the safety information of the unknown drug combination, providing a data base for the unknown drug combination and reducing the risk of the unknown drug combination.

For example, one of the scenarios in the embodiment of the present application may be applied to the scenario shown in fig. 1, which includes the drug combination safety evaluation system 100. The user inputs the drugs to be combined into the drug combination safety evaluation system 100, and the drug combination safety evaluation system 100 performs drug combination safety evaluation according to the drugs to be combined to obtain the safety evaluation result of the drugs to be combined, so that the user can know the safety evaluation result of the drugs to be combined.

First, in the application scenario, although the actions of the embodiment provided in the embodiment of the present application are described as being performed by the drug combination safety evaluation system 100; however, the embodiments of the present application are not limited in terms of executing subjects as long as the actions disclosed in the embodiments provided by the embodiments of the present application are executed.

Next, the above scenario is only one example of the scenario provided in the embodiment of the present application, and the embodiment of the present application is not limited to this scenario.

The following describes in detail specific implementations of the system, method and device for evaluating drug combination safety according to embodiments of the present application, with reference to the accompanying drawings.

Exemplary System

Referring to fig. 2, a schematic structural diagram of a drug combination safety evaluation system in an embodiment of the present application is shown. In this embodiment, the system may specifically include:

an input module 201, a subject database 202 and a security evaluation module 203;

the input module 201 is used for obtaining a first medicament and a second medicament to be combined;

the topic database 202 is configured to determine each chemical component in the first drug and each chemical component in the second drug according to the first drug and the second drug;

the safety evaluation module 203 is configured to obtain a multi-organ damage prediction result of each chemical component in the first drug and a multi-organ damage prediction result of each chemical component in the second drug according to each chemical component in the first drug, each chemical component in the second drug and a drug combination multi-organ damage prediction model, and use the multi-organ damage prediction results as safety evaluation results of the first drug and the second drug to be combined;

In the prior art, users such as clinicians and medical researchers obtain various chemical components and corresponding multi-organ injury results of combined medicines through published literature data, network information, medical books and the like, and for unknown combined medicines, whether the unknown combined medicines cause multi-organ injury cannot be known. Therefore, in the embodiment of the application, a drug combination safety evaluation system comprising an input module, a topic database and a safety evaluation module is considered to be constructed, and the drug combination safety evaluation system is used for predicting whether chemical components in unknown drug combinations cause multi-organ damage.

During specific implementation, a user inputs the drugs to be combined, such as a first drug and a second drug, in the drug combination safety evaluation system, and firstly, the drug combination safety evaluation system can obtain the first drug and the second drug to be combined through the input module; then, the essence of the first medicament and the second medicament needs to be determined, namely, each chemical component in the first medicament and each chemical component in the second medicament need to be determined, each chemical component in each medicament can be recorded in a theme database in advance, and each chemical component in the first medicament and each chemical component in the second medicament can be determined by the medicament combined safety evaluation system through the theme database; and finally, pre-training to obtain a drug combination multi-organ injury prediction model through each chemical component in the drugs and the corresponding multi-organ injury result, storing the drug combination multi-organ injury prediction model in a safety evaluation module, and on the basis of the data, obtaining the multi-organ injury prediction result of each chemical component in the first drug and the multi-organ injury prediction result of each chemical component in the second drug through the safety evaluation module by the drug combination safety evaluation system to serve as the safety evaluation results of the first drug and the second drug to be combined.

Wherein, the training of the drug combination multi-organ injury prediction model is as follows: taking each chemical component in the medicine and the corresponding multiple organ injury result as training data; training a neural network through the training data to fully mine the correlation information between chemical components in the study medicine and the corresponding multi-organ injury results; after a certain amount of training data are subjected to iterative training for multiple times, a drug-combined multi-organ injury prediction model can be obtained.

In the embodiment of the application, the research focus of users such as clinicians and medical researchers is on the combination of Chinese and western medicines, and the first medicine and the second medicine to be combined can be the Chinese medicine and the western medicine to be combined; correspondingly, when the theme database 202 is used for determining each chemical component in the traditional Chinese medicine and each chemical component in the western medicine, the traditional Chinese medicine is considered to be more complex than the western medicine; for traditional Chinese medicine, the subject database 202 needs to provide a traditional Chinese medicine basic database, a traditional Chinese medicine compound basic database and a traditional Chinese medicine chemical composition database; for western medicines, the topic database 202 may provide a database of chemical components of the western medicines. In an optional implementation manner of the embodiment of the present application, the first medicine is a traditional Chinese medicine, the second medicine is a western medicine, and the topic database 202 includes a traditional Chinese medicine basic database, a traditional Chinese medicine compound basic database, a traditional Chinese medicine chemical component database and a western medicine chemical component database.

The Chinese medicine basic database comprises 14 fields of Chinese medicine names, Chinese medicine names-Latin, indications-Chinese, liver injury, heart injury, lung injury, kidney injury, brain injury, chemical components, Chinese medicine compound containing the Chinese medicines, target genes, related diseases, adverse reactions and references. The basic database of the Chinese herbal compound comprises 6 fields of compound name-Pinyin, compound name-Chinese, Chinese herbal compound formulation, Chinese herbal medicine-Pinyin in compound, Chinese herbal medicine-Chinese in compound, and indication-Chinese. The Chinese medicine Chemical component database comprises 25 fields of Chemical component names, Chemical structural formulas, Molecular formulas, organic small molecule bioactivity data (English: Puchem) numbers, American Chemical Abstracts Service (CAS) registration numbers, Molecular weights, Simplified Molecular linear Input specifications (English: Simplified Molecular Input Line Entry System (SMILES) codes, hydrogen bond receptors, hydrogen bond donors, rotatable bonds, water solubility, oil-water partition coefficient (LogP values), liver injury, heart injury, lung injury, kidney injury, brain injury, Chinese medicines containing the Chemical components, Chinese medicine compounds containing the Chemical components, target genes, passages, related diseases, adverse events, other documents and reference documents. The database of western medicine chemical composition includes 25 fields for western medicine name, indication, related therapy, action mechanism, chemical structural formula, molecular formula, CAS registry number, molecular weight, SMILES code, protein binding, absorption, volume of distribution, metabolism, elimination pathway, drug interaction, liver injury, heart injury, lung injury, kidney injury, brain injury, target gene, related disease, adverse reaction, others and references.

For each chemical component in the drug, the SMILES code of the chemical component is a linear symbol for inputting and representing molecular reactions, and is an artificial intelligence and chemical expert's language-ASCII code; the theme database 202 is used to determine that each chemical component in the traditional Chinese medicine and each chemical component in the western medicine actually means: determining SMILES codes of all chemical components in the traditional Chinese medicine and SMILES codes of all chemical components in the western medicine. Specifically, the theme database 202 for determining each chemical component in the traditional Chinese medicine means: firstly, inquiring each chemical component in the traditional Chinese medicine in a traditional Chinese medicine basic database or a traditional Chinese medicine compound basic database; and then inquiring SMILES codes of all chemical components in the traditional Chinese medicine in a traditional Chinese medicine chemical component database. The theme database 202 is used for determining each chemical component in the western medicine, and the chemical components refer to: and inquiring the SMILES codes of all chemical components in the western medicine chemical component database.

That is, in an optional implementation manner of this embodiment of the present application, the theme database 202 is specifically configured to query the traditional Chinese medicine basic database or the traditional Chinese medicine prescription basic database according to the traditional Chinese medicine, determine each chemical component in the traditional Chinese medicine, query the traditional Chinese medicine chemical component database according to each chemical component in the traditional Chinese medicine, and determine a simplified molecular linear input specification SMILES code of each chemical component in the traditional Chinese medicine; and querying the chemical component database of the western medicine according to the western medicine to determine the SMILES of each chemical component in the western medicine.

Correspondingly, the drug combination multi-organ damage prediction model comprises an embedded layer and a prediction network, and on the basis of the SMILES code, the actual process of obtaining the safety evaluation result of the traditional Chinese medicine and the western medicine to be combined by the safety evaluation module 203 is as follows: firstly, inputting SMILES codes of all chemical components in the traditional Chinese medicine and SMILES codes of all chemical components in the western medicine into an embedding layer in a medicine combination multi-organ injury prediction model, wherein the embedding layer is used for vectorizing compound molecules of all chemical components in the traditional Chinese medicine and all chemical components in the western medicine through the SMILES codes to obtain characteristic vectors of all chemical components in the traditional Chinese medicine and characteristic vectors of all chemical components in the western medicine; then, on the basis of the characteristic vectors, inputting the characteristic vectors of all chemical components in the traditional Chinese medicine and the characteristic vectors of all chemical components in the western medicine into a prediction network in a medicine combined multi-organ damage prediction model, predicting whether all chemical components in the traditional Chinese medicine cause multi-organ damage or not, obtaining multi-organ damage prediction results of all chemical components in the traditional Chinese medicine and multi-organ damage prediction results of all chemical components in the western medicine, wherein the results are safety evaluation results of the traditional Chinese medicine and the western medicine to be combined.

That is, in an optional implementation manner of the embodiment of the present application, the safety evaluation module 203 is specifically configured to obtain feature vectors of each chemical component in the traditional Chinese medicine and feature vectors of each chemical component in the western medicine according to the SMILES code of each chemical component in the traditional Chinese medicine, the SMILES code of each chemical component in the western medicine, and the embedded layer in the multi-organ damage prediction model for pharmaceutical combination; and obtaining a multi-organ damage prediction result of each chemical component in the traditional Chinese medicine and a multi-organ damage prediction result of each chemical component in the western medicine according to the feature vectors of each chemical component in the traditional Chinese medicine, the feature vectors of each chemical component in the western medicine and a prediction network in the medicine combined multi-organ damage prediction model, and taking the multi-organ damage prediction results as safety evaluation results of the traditional Chinese medicine and the western medicine to be combined.

In addition, in the embodiment of the application, in the process that the safety evaluation module 203 obtains the multi-organ damage prediction result of each chemical component in the traditional Chinese medicine and the multi-organ damage prediction result of each chemical component in the western medicine, each chemical component in the traditional Chinese medicine and each chemical component in the western medicine can interact with each other under a certain metabolic enzyme to generate certain metabolites, and whether the metabolites cause multi-organ damage or not needs to be determined, that is, whether the metabolites have safety or not; therefore, the safety evaluation module 203 is further configured to combine preset metabolic enzymes as preset metabolic enzymes on the basis of each chemical component in the traditional Chinese medicine and each chemical component in the western medicine, wherein the preset metabolic enzymes are 5 metabolic enzymes such as CYP1a2, CYP2C9, CYP3a4, CYP2D6 or CYP2C19, and the like, input the drug combination multi-organ injury prediction model, predict the generated metabolites, and obtain each predicted metabolite of each chemical component in the traditional Chinese medicine and each predicted metabolite of each chemical component in the western medicine interacting; based on the above, on the basis of each predicted metabolite, a drug combined multi-organ injury prediction model is used for predicting whether each predicted metabolite causes multi-organ injury, so as to obtain a multi-organ injury prediction result of each predicted metabolite, and the multi-organ injury prediction result is also used as a safety evaluation result of a traditional Chinese medicine and a western medicine to be combined.

That is, in an optional implementation manner of the embodiment of the present application, the safety evaluation module 203 is further configured to obtain, according to each chemical component in the traditional Chinese medicine, each chemical component in the western medicine, a preset metabolic enzyme, and the drug combined multi-organ injury prediction model, each predicted metabolite of interaction between each chemical component in the traditional Chinese medicine and each chemical component in the western medicine; and obtaining a multi-organ injury prediction result of each predicted metabolite according to the multi-organ injury prediction model combining each predicted metabolite and the medicine, and taking the multi-organ injury prediction result as a safety evaluation result of the traditional Chinese medicine and the western medicine to be combined.

Therefore, the model for predicting the damage of the multi-organ caused by drug combination in the safety evaluation module can also predict whether the damage of the multi-organ caused by the metabolite generated by the interaction of chemical components in the unknown drug combination is caused, so that the safety information of the unknown drug combination is further determined, a data base is further provided for the unknown drug combination, and the risk of the unknown drug combination is reduced.

Referring to the above description of the SMILES code, the obtaining of the actual processes of each predicted metabolite by the safety evaluation module 203 means: and obtaining each predicted metabolite of each chemical component in the traditional Chinese medicine and each chemical component in the western medicine, which interact with each other, according to the feature vector of each chemical component in the traditional Chinese medicine, the feature vector of each chemical component in the western medicine, a preset metabolic enzyme and a prediction network in the medicine combined multi-organ injury prediction model.

Referring again to the above description of the SMILES code, for each predicted metabolite, it is also necessary to determine the SMILES code for each predicted metabolite, typically using a pre-programmed chemical editor. The actual process of obtaining the multiple organ injury prediction results of the predicted metabolites by the safety evaluation module 203 is as follows: firstly, inputting SMILES codes of all predicted metabolites into an embedding layer in a drug combination multi-organ injury prediction model, wherein the embedding layer is used for vectorizing compound molecules of all predicted metabolites and the like through the SMILES codes to obtain characteristic vectors of all predicted metabolites; and then, on the basis of the characteristic vectors, inputting the characteristic vectors of the predicted metabolites into a prediction network in a drug combination multi-organ injury prediction model, predicting whether the predicted metabolites cause multi-organ injury or not, and obtaining multi-organ injury prediction results of the predicted metabolites.

That is, in an optional implementation manner of the embodiment of the present application, the safety evaluation module 203 is further specifically configured to obtain, by using a preset chemical editor, a SMILES code of each predicted metabolite, and obtain a feature vector of each predicted metabolite according to the SMILES code of each predicted metabolite and an embedded layer in the drug-combination multi-organ injury prediction model; and obtaining the multi-organ injury prediction result of each predicted metabolite according to the feature vector of each predicted metabolite and a prediction network in the drug combined multi-organ injury prediction model.

In the embodiment of the application, the safety information of the combined medicines such as each chemical component in the combined medicines and the corresponding injury result of multiple organs is obtained by considering the fact that users such as clinicians, medical researchers and the like need to respectively obtain the safety information of the combined medicines such as published literature data, network information, medical books and the like; the drug combination safety information is dispersed and not centralized, the information is not comprehensive enough, a certain time is needed for collecting the information, and the information can be collected in advance and stored in the theme database 202. For the research focus of users such as clinicians and medical researchers and the like on the combination of Chinese and western medicines, the subject database 202 also needs to provide a clinical database of the combination of Chinese and western medicines, a medicine interaction database in the combination of Chinese and western medicines, a metabolic reaction database of the combination of Chinese and western medicines, a disease database of the combination of Chinese and western medicines, a multi-organ drug injury database and a multi-organ injury related pathway database; the system for evaluating the safety of the combination of the medicines also needs to provide a search module, wherein the search module at least comprises a combined Chinese and western medicine search submodule, so that a user can search the database through the combined Chinese and western medicine search submodule.

That is, in an optional implementation manner of this embodiment of the present application, the subject database 202 further includes a combined chinese and western medicine clinical database, a drug interaction database in combined chinese and western medicine, a combined chinese and western medicine metabolic reaction database, a combined chinese and western medicine disease database, a multi-organ drug injury database, and a multi-organ injury related pathway database; correspondingly, the system further comprises a search module, the search module at least comprises a combined Chinese and Western medicine search submodule, and the combined Chinese and Western medicine search submodule is used for searching the clinical database of combined Chinese and Western medicine, the drug interaction database of combined Chinese and Western medicine, the metabolic reaction database of combined Chinese and Western medicine, the disease database of combined Chinese and Western medicine, the multi-organ drug injury database and the multi-organ injury related pathway database according to the traditional Chinese medicine and the western medicine.

The clinical database of the combination of Chinese and western medicines comprises 9 fields of combination of Chinese and western medicines, indications, combination advantages, risk factors, multiple organ injuries, clinical safety and effectiveness, other adverse reactions, clinical reports and reference documents. The medicine interaction database in the combination of Chinese and western medicines comprises 8 fields of the combination of Chinese and western medicines, absorption interaction, neutralization reaction/complexation, metabolism interaction, distribution interaction, excretion interaction, adverse reaction and reference documents. The database of the metabolic reaction of the combination of Chinese and Western medicines comprises 7 fields of combination of Chinese and Western medicines, substrates, metabolic enzymes, other enzymes, metabolic reactions, adverse reactions and references. The disease database of the combination of Chinese and western medicines comprises 7 fields of combination of Chinese and western medicines, diseases, disease categories, dosage forms, combination advantages, mechanisms and reference documents. The multi-organ drug-induced injury database comprises 25 fields including name, chemical structural formula, molecular formula, Puchem number, CAS registry number, molecular weight, SMILES code, hydrogen bond acceptor, hydrogen bond donor, rotatable bond, water solubility, LogP value, liver injury, heart injury, lung injury, kidney injury, brain injury, Chinese medicine containing the chemical component, Chinese medicine compound containing the chemical component, target gene, pathway, related disease, adverse event, others and reference. The multi-organ injury-associated pathway database comprises 6 fields of organ injury, pathway ID, chemical composition, target gene and reference.

In addition, in this embodiment of the application, the search module may further provide one or more of a chinese medicine search sub-module for searching related information of a chinese medicine, a chemical structural formula search sub-module for searching related information of a chemical structural formula, an advanced search sub-module for implementing advanced search, and the like. In an optional implementation manner of the embodiment of the present application, the search module further includes a chinese medicine search submodule, a chemical structural formula search submodule, and/or an advanced search submodule.

As an example, as shown in fig. 3, a schematic structure diagram of another drug combination safety evaluating system is shown, and compared with the drug combination safety evaluating system shown in fig. 2, a search module 204 is added to the drug combination safety evaluating system shown in fig. 3.

In addition, in this embodiment of the application, the drug combination safety evaluation system further needs to display the safety evaluation result of the first drug and the second drug to be combined, which is obtained by the safety evaluation module 203, for users such as clinicians and medical researchers, and the visualization module is further needed to be provided by the drug combination safety evaluation system, so that after the safety evaluation module 203 obtains the safety evaluation result of the first drug and the second drug to be combined, the safety evaluation result of the first drug and the second drug to be combined is visually displayed. That is, in an optional implementation manner of the embodiment of the present application, the system further includes a visualization module, and the visualization module is configured to visually display a safety evaluation result of the first drug and the second drug to be combined.

As another example, as shown in fig. 4, a schematic structure diagram of another drug combination safety evaluation system is shown, and compared with the drug combination safety evaluation system shown in fig. 3, a visualization module 205 is added to the drug combination safety evaluation system shown in fig. 4.

Based on the above description, the user can input the traditional Chinese medicine and the western medicine to be combined in the drug combination safety evaluation system, and when the drug combination safety evaluation system is in an English version, the user needs to input the traditional Chinese medicine-English and the western medicine-English, and related prompts exist; when the drug combination safety evaluation system is in a Chinese version, Chinese medicines-Chinese and western medicines-Chinese need to be input, and fuzzy prompt is given. Correspondingly, the step of visually displaying the safety evaluation result of the first medicament and the second medicament to be combined is as follows: the multi-organ injury prediction results of all chemical components in the traditional Chinese medicine are displayed in a combined and visual mode; predicting the multi-organ damage of each chemical component in western medicine; each predicted metabolite of interaction of each chemical component in the traditional Chinese medicine and each chemical component in the western medicine; and multiple organ injury predictions for each predicted metabolite, and so forth. For example, a combination of traditional Chinese medicine and western medicine predicts chemical components (prototype compounds), metabolic reactions, corresponding metabolites, etc. at risk of liver injury (heart injury, brain injury, kidney injury, or lung injury). In addition, the risk evaluation value obtained by calculation by using a preset formula based on the safety evaluation result of the first medicine and the second medicine to be combined can be visually displayed. For example, see the following example of the safety evaluation report of the combination of Chinese and western medicines.

The sample of the safety evaluation report of the combined use of the traditional Chinese medicine and the western medicine comprises the following steps:

prototype compounds that predict the presence of a risk of liver injury are:

correspondingly, inhibitors, inducers, activators of metabolic enzymes are predicted to be present:

the compounds and reaction types for which metabolic reactions are predicted are:

prototype compounds that are predicted to be at risk of heart injury are:

prototype compounds that are predicted to be at risk of brain injury are:

prototype compounds that predict the presence of a risk of renal injury are:

prototype compounds predicted to be at risk of lung injury are:

overall, the risk assessment values (scores or grades) are: and (4) X.

Of course, the visualization module is also used for visually displaying metabolic reaction, passage, chemical structural formula and disease-related target points. For example, the metabolic reaction process is visualized by using a chemical information toolkit-RDkit or the like according to each predicted metabolite; visually displaying the path of the target point of the first medicament and the target point of the second medicament to be combined through a visual toolkit-pathview and the like; visually displaying the chemical structural formulas of the first medicament and the second medicament to be combined by adopting a chemical structure visualization tool, namely LiteMol and the like; calling a corresponding target file in a subject database, rendering the three-dimensional structure of the target protein, displaying the structure of the corresponding ligand compound, listing the detailed conditions of the ligand compound, and adopting target protein structure display tools-NGL, 3Dmol, Litelol and the like.

According to various embodiments provided by the embodiment, the drug combination safety evaluation system comprises an input module, a theme database and a safety evaluation module; the input module is used for obtaining a first medicament and a second medicament to be combined; the theme database is used for determining each chemical component in the first medicament and each chemical component in the second medicament according to the first medicament and the second medicament; the safety evaluation module is used for obtaining a multi-organ damage prediction result of each chemical component in the first medicine and a multi-organ damage prediction result of each chemical component in the second medicine according to each chemical component in the first medicine, each chemical component in the second medicine and the medicine combination multi-organ damage prediction model, and using the multi-organ damage prediction results as safety evaluation results of the first medicine and the second medicine to be combined; the drug combination multi-organ injury prediction model is obtained by pre-training a neural network according to each chemical component in the drug and a corresponding multi-organ injury result.

Therefore, the drug combination safety evaluation system firstly determines each chemical component in the drugs to be combined through the theme database, then obtains the multi-organ damage prediction result of each chemical component through the drug combination multi-organ damage prediction model for fully learning the incidence relation between each chemical component in the drugs and the corresponding multi-organ damage result in the safety evaluation module, and realizes prediction on whether the chemical components in the unknown drug combination cause multi-organ damage, so that the unknown drug combination safety information is determined, a data base is provided for the unknown drug combination, and the risk of the unknown drug combination is reduced.

Exemplary method

Referring to fig. 5, a schematic flow chart of a method for evaluating the safety of drug combination according to an embodiment of the present application is shown. In this embodiment, the method may include, for example, the steps of:

step 501: obtaining a first drug and a second drug to be combined.

Step 502: determining respective chemical components in the first medicament and respective chemical components in the second medicament based on the first medicament and the second medicament.

Step 503: and obtaining a multi-organ injury prediction result of each chemical component in the first medicament and a multi-organ injury prediction result of each chemical component in the second medicament according to each chemical component in the first medicament, each chemical component in the second medicament and a medicament combined multi-organ injury prediction model, and taking the multi-organ injury prediction results as safety evaluation results of the first medicament and the second medicament to be combined.

In an optional implementation manner of the embodiment of the present application, the first medicine is a traditional Chinese medicine, and the second medicine is a western medicine.

In an optional implementation manner of this embodiment of this application, the step 502 may include the following steps:

step A: inquiring a traditional Chinese medicine basic database or a traditional Chinese medicine compound basic database according to the traditional Chinese medicine, and determining each chemical component in the traditional Chinese medicine;

and B: querying a traditional Chinese medicine chemical component database according to each chemical component in the traditional Chinese medicine, and determining a simplified molecular linear input standard SMILES code of each chemical component in the traditional Chinese medicine;

and C: querying a western medicine chemical component database according to the western medicine to determine SMILES (small entry loss) codes of all chemical components in the western medicine;

correspondingly, the step 503 may include the following steps, for example:

step D: obtaining a feature vector of each chemical component in the traditional Chinese medicine and a feature vector of each chemical component in the western medicine according to the SMILES codes of each chemical component in the traditional Chinese medicine, the SMILES codes of each chemical component in the western medicine and the embedded layer in the medicine combined multi-organ injury prediction model;

step E: and obtaining a multi-organ damage prediction result of each chemical component in the traditional Chinese medicine and a multi-organ damage prediction result of each chemical component in the western medicine according to the feature vectors of each chemical component in the traditional Chinese medicine, the feature vectors of each chemical component in the western medicine and a prediction network in the medicine combined multi-organ damage prediction model, and taking the multi-organ damage prediction results as safety evaluation results of the traditional Chinese medicine and the western medicine to be combined.

In an optional implementation manner of the embodiment of the present application, for example, the following steps may be further included:

step F: obtaining each predicted metabolite of interaction of each chemical component in the traditional Chinese medicine and each chemical component in the western medicine according to each chemical component in the traditional Chinese medicine, each chemical component in the western medicine, a preset metabolic enzyme and the medicine combined multi-organ injury prediction model;

step G: and obtaining a multi-organ injury prediction result of each predicted metabolite according to the multi-organ injury prediction model combining each predicted metabolite and the medicine, and taking the multi-organ injury prediction result as a safety evaluation result of the traditional Chinese medicine and the western medicine to be combined.

In an optional implementation manner of the embodiment of the present application, the step G may include, for example, the following steps:

step G1: obtaining SMILES codes of the predicted metabolites by using a preset chemical editor, and obtaining feature vectors of the predicted metabolites according to the SMILES codes of the predicted metabolites and an embedded layer in the drug combination multi-organ injury prediction model;

step G2: and obtaining the multi-organ injury prediction result of each predicted metabolite according to the feature vector of each predicted metabolite and a prediction network in the drug combined multi-organ injury prediction model.

searching a Chinese and western medicine combined clinical database, a medicine interaction database in the combination of Chinese and western medicines, a Chinese and western medicine combined metabolic reaction database, a Chinese and western medicine combined disease database, a multi-organ drug injury database and a multi-organ injury related pathway database according to the traditional Chinese medicines and the western medicines.

In an optional implementation manner of the embodiment of the present application, the search manner includes chinese medicine search, chemical formula search, and/or advanced search.

In an optional implementation manner of the embodiment of the present application, for example, the following steps may be further included: and visually displaying the safety evaluation result of the first medicine and the second medicine to be combined.

By the various embodiments provided in this example, a first drug and a second drug to be combined are obtained; determining each chemical component in the first medicament and each chemical component in the second medicament according to the first medicament and the second medicament; according to each chemical component in the first medicine, each chemical component in the second medicine and the medicine combination multi-organ injury prediction model, obtaining a multi-organ injury prediction result of each chemical component in the first medicine and a multi-organ injury prediction result of each chemical component in the second medicine, and taking the multi-organ injury prediction results as safety evaluation results of the first medicine and the second medicine to be combined; the drug combination multi-organ injury prediction model is obtained by pre-training a neural network according to each chemical component in the drug and a corresponding multi-organ injury result. Therefore, each chemical component in the medicines to be combined is determined firstly, then the multi-organ damage prediction model of the incidence relation between each chemical component in the medicines and the corresponding multi-organ damage result is fully learned to obtain the multi-organ damage prediction result of each chemical component, and whether the chemical components in the unknown medicines are damaged by multiple organs or not is predicted, so that the safety information of the unknown medicines is determined, a data basis is provided for the unknown medicines to be combined, and the risk of the unknown medicines to be combined is reduced.

Exemplary devices

Referring to fig. 6, a schematic structural diagram of a drug combination safety evaluation device in an embodiment of the present application is shown. In this embodiment, the apparatus may specifically include:

a first obtaining unit 601 for obtaining a first drug and a second drug to be combined;

a determining unit 602, configured to determine, according to the first drug and the second drug, respective chemical components in the first drug and respective chemical components in the second drug;

a second obtaining unit 603, configured to obtain a multiple organ injury prediction result of each chemical component in the first drug and a multiple organ injury prediction result of each chemical component in the second drug according to each chemical component in the first drug, each chemical component in the second drug, and a drug combination multiple organ injury prediction model, as a safety evaluation result of the first drug and the second drug to be combined;

In an optional implementation manner of the embodiment of the present application, the determining unit 602 may include:

the first determining subunit is used for inquiring a traditional Chinese medicine basic database or a traditional Chinese medicine compound basic database according to the traditional Chinese medicine and determining each chemical component in the traditional Chinese medicine;

the second determining subunit is used for querying a traditional Chinese medicine chemical component database according to each chemical component in the traditional Chinese medicine and determining a simplified molecular linear input specification SMILES code of each chemical component in the traditional Chinese medicine;

a third determining subunit, configured to query a western medicine chemical component database according to the western medicine to determine a SMILES code of each chemical component in the western medicine;

correspondingly, the second obtaining unit 603 may include, for example:

a first obtaining subunit, configured to obtain feature vectors of each chemical component in the traditional Chinese medicine and feature vectors of each chemical component in the western medicine according to the SMILES code of each chemical component in the traditional Chinese medicine, the SMILES code of each chemical component in the western medicine, and the embedded layer in the drug combination multi-organ injury prediction model;

and the second obtaining subunit is used for obtaining a multi-organ damage prediction result of each chemical component in the traditional Chinese medicine and a multi-organ damage prediction result of each chemical component in the western medicine according to the feature vectors of each chemical component in the traditional Chinese medicine, the feature vectors of each chemical component in the western medicine and the prediction network in the medicine combined multi-organ damage prediction model, and using the multi-organ damage prediction results as the safety evaluation results of the traditional Chinese medicine and the western medicine to be combined.

In an optional implementation manner of the embodiment of the present application, the apparatus may further include:

a third obtaining unit, configured to obtain predicted metabolites of interaction between each chemical component in the traditional Chinese medicine and each chemical component in the western medicine according to each chemical component in the traditional Chinese medicine, each chemical component in the western medicine, a preset metabolic enzyme, and the medicine combined multi-organ injury prediction model;

and a fourth obtaining unit, configured to obtain a multiple organ injury prediction result of each predicted metabolite according to the multiple organ injury prediction model combining each predicted metabolite and the drug, where the multiple organ injury prediction result is used as a safety evaluation result of the traditional Chinese medicine and the western medicine to be combined.

In an optional implementation manner of the embodiment of the present application, the fourth obtaining unit may include, for example:

a third obtaining subunit, configured to obtain, by using a preset chemical editor, the SMILES code of each predicted metabolite, and obtain, according to the SMILES code of each predicted metabolite and an embedded layer in the drug-coupled multi-organ injury prediction model, a feature vector of each predicted metabolite;

and the fourth obtaining subunit is used for obtaining the multi-organ injury prediction result of each predicted metabolite according to the feature vector of each predicted metabolite and the prediction network in the drug combination multi-organ injury prediction model.

and the searching unit is used for searching a Chinese and western medicine combined clinical database, a medicine interaction database in combined Chinese and western medicine, a Chinese and western medicine combined metabolic reaction database, a Chinese and western medicine combined disease database, a multi-organ drug injury database and a multi-organ injury related pathway database according to the traditional Chinese medicines and the western medicines.

and the visual display unit is used for visually displaying the safety evaluation result of the first medicine and the second medicine to be combined.

The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.

Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The foregoing is illustrative of the preferred embodiments of the present application and is not to be construed as limiting the present application in any way. Although the present application has been described with reference to the preferred embodiments, it is not intended to limit the present application. Those skilled in the art can now make numerous possible variations and modifications to the disclosed embodiments, or modify equivalent embodiments, using the methods and techniques disclosed above, without departing from the scope of the claimed embodiments. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application still fall within the protection scope of the technical solution of the present application without departing from the content of the technical solution of the present application.

Claims

1. A drug combination safety evaluation system is characterized by comprising: the system comprises an input module, a theme database and a safety evaluation module;

2. The system of claim 1, wherein the first medicine is a traditional Chinese medicine, the second medicine is a western medicine, and the subject database comprises a traditional Chinese medicine basic database, a traditional Chinese medicine compound basic database, a traditional Chinese medicine chemical composition database and a western medicine chemical composition database.

3. The system of claim 2, wherein the topic database is specifically configured to query the primary database of chinese herbs or the primary database of chinese herbal compound according to the chinese herbs to determine each chemical component in the chinese herbs, query the database of chemical components in the chinese herbs according to each chemical component in the chinese herbs to determine a simplified molecular linear input specification SMILES code for each chemical component in the chinese herbs; querying the chemical component database of the western medicine according to the western medicine to determine SMILES (small entry loss) codes of all chemical components in the western medicine;

4. The system according to claim 2, wherein the safety evaluation module is further configured to obtain predicted metabolites of interaction of each chemical component in the traditional Chinese medicine and each chemical component in the western medicine according to each chemical component in the traditional Chinese medicine, each chemical component in the western medicine, a preset metabolic enzyme and the drug combination multi-organ injury prediction model; and obtaining a multi-organ injury prediction result of each predicted metabolite according to the multi-organ injury prediction model combining each predicted metabolite and the medicine, and taking the multi-organ injury prediction result as a safety evaluation result of the traditional Chinese medicine and the western medicine to be combined.

5. The system according to claim 4, wherein the safety evaluation module is further specifically configured to obtain a SMILES code of each of the predicted metabolites using a pre-programmed chemical editor, and obtain a feature vector of each of the predicted metabolites according to the SMILES code of each of the predicted metabolites and an embedded layer in the drug combination multi-organ injury prediction model; and obtaining the multi-organ injury prediction result of each predicted metabolite according to the feature vector of each predicted metabolite and a prediction network in the drug combined multi-organ injury prediction model.

6. The system of claim 2, wherein said subject database further comprises a combined chinese and western clinical database, a drug interaction database in combined chinese and western, a combined chinese and western metabolic response database, a combined chinese and western disease database, a multi-organ drug injury database, and a multi-organ injury-related pathway database; correspondingly, the system further comprises a search module, the search module at least comprises a combined Chinese and Western medicine search submodule, and the combined Chinese and Western medicine search submodule is used for searching the clinical database of combined Chinese and Western medicine, the drug interaction database of combined Chinese and Western medicine, the metabolic reaction database of combined Chinese and Western medicine, the disease database of combined Chinese and Western medicine, the multi-organ drug injury database and the multi-organ injury related pathway database according to the traditional Chinese medicine and the western medicine.

7. The system of claim 6, wherein the search module further comprises a chinese medicine search sub-module, a chemical formula search sub-module, and/or an advanced search sub-module.

8. The system according to claim 1, further comprising a visualization module for visually displaying the result of the safety assessment of the first and second drugs to be combined.

9. A drug combination safety evaluation method is characterized by comprising the following steps:

obtaining a first drug and a second drug to be combined;

10. A drug combination safety evaluation device, comprising: