Knowledge graph-based terpene compound and disease correlation prediction method and system

Abstract

The invention relates to the technical field of artificial intelligence and pharmacology, and provides a terpene compound and disease correlation prediction method and system based on a mapping knowledge domain, and the method comprises the following steps: collecting terpene compounds and biological activity data thereof, including terpene compound molecular structure, protein target, gene target, cell line and disease correlation information, constructing a terpenoid biological activity knowledge graph; inputting the molecular descriptor of the to-be-predicted terpenoid into a convolutional neural network for feature extraction to generate a first feature vector; performing feature extraction on the molecular structure of the to-be-predicted terpenoid through a graph convolutional network to generate a second feature vector; a molecular embedding vector obtained by splicing the first feature vector and the second feature vector is embedded into a model through a knowledge graph, and a terpenoid-disease prediction score matrix is generated; and performing descending sorting based on the terpenoids-disease prediction scores, and generating a disease recommendation list related to the terpenoids.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and pharmacology technology, and more specifically, to a method and system for predicting the correlation between terpenoid compounds and diseases based on a knowledge graph. Background Art

[0002] Terpenoids, as natural compounds widely found in plants and other organisms, have attracted widespread attention in the field of drug development due to their diverse structures and wide range of biological activities. Terpenoids show potential in a variety of pharmacological activities such as anti-inflammatory, anti-tumor, and antimicrobial. However, the confirmation of the pharmacological activity of terpenoids mainly relies on traditional experimental methods, including in vitro cell experiments, animal experiments, and clinical trials. Although these experimental methods are widely recognized and technically feasible, they are often very time-consuming and accompanied by high resource and time investment, which limits the efficiency of new drug discovery and development.

[0003] As the application of artificial intelligence in drug development gradually matures, algorithms based on machine learning and deep learning can quickly process and analyze large amounts of data, significantly accelerating the process of drug activity screening, drug target discovery, and drug-disease correlation prediction. It is currently proposed to build a knowledge graph dataset centered on drugs and targets, and use artificial intelligence models to learn the characteristic information of drugs, targets, and their neighborhoods to quickly predict the potential therapeutic effect of a terpene compound on a specific disease, thereby reducing the time and cost of experimental verification. This type of model can extract the commonalities and laws based on the existing association data of terpene compounds and diseases, and apply them to the drug development of new terpene compounds, providing direction for future drug development. However, there is currently a lack of multidimensional relationship network data that comprehensively covers the activity profile of terpenes and prediction models for potential diseases, which leads to limitations in disease-related prediction results in specific scenarios. Summary of the invention

[0004] In order to overcome the defect of the above-mentioned prior art that the data source is single, resulting in limitations in disease-related prediction results in specific scenarios, the present invention provides a terpenoid compound and disease correlation prediction method and system based on a knowledge graph.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] A method for predicting the correlation between terpenoid compounds and diseases based on a knowledge graph comprises the following steps:

[0007] Collect data on terpenoids and their biological activities, including molecular structures, protein targets, gene targets, cell lines, and disease association information of terpenoids, and construct a knowledge graph of terpenoid biological activities;

[0008] The molecular descriptors of the terpenoid compounds to be predicted are input into a convolutional neural network for feature extraction to generate a first feature vector about the physical and chemical properties of the molecules;

[0009] The molecular structure of the terpenoid compound to be predicted is subjected to feature extraction through a graph convolutional network to generate a second feature vector regarding the features of the terpenoid molecular graph;

[0010] The molecular embedding vector obtained by concatenating the first feature vector and the second feature vector is embedded in the knowledge graph model to generate a terpenoid compound-disease prediction score matrix;

[0011] Based on the terpenoid-disease prediction scores, the disease recommendation list related to terpenoids was generated by descending sorting.

[0012] Furthermore, the present invention also proposes a terpenoid compound and disease correlation prediction system based on knowledge graph, and applies the terpenoid compound and disease correlation prediction method proposed in the present invention. The system comprises:

[0013] The knowledge graph construction module is used to collect data on terpenoid compounds and their biological activities, including terpenoid molecular structures, protein targets, gene targets, cell lines and disease association information, and to construct a knowledge graph of terpenoid biological activities;

[0014] A first feature extraction module is used to input the molecular descriptors of the terpenoid compounds to be predicted into the convolutional neural network for feature extraction, and generate a first feature vector about the physical and chemical properties of the molecules;

[0015] A second feature extraction module is used to extract features of the molecular structure of the terpenoid compound to be predicted through a graph convolutional network to generate a second feature vector about the features of the terpenoid molecular graph;

[0016] A knowledge graph embedding module, used to embed the molecular embedding vector obtained by concatenating the first feature vector and the second feature vector through a knowledge graph embedding model to generate a terpenoid compound-disease prediction score matrix;

[0017] The output module is used to generate a recommended list of diseases related to terpenoids by sorting them in descending order based on the terpenoid-disease prediction scores.

[0018] Furthermore, the present invention also proposes a device, comprising a memory and a processor, wherein the memory stores computer-readable instructions, wherein when the computer-readable instructions are executed by the processor, the processor executes all or part of the steps of the knowledge graph-based terpene compound and disease correlation prediction method as described in the present invention.

[0019] Furthermore, the present invention also proposes a storage medium having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by a processor, implement all or part of the steps of the knowledge graph-based terpene compound and disease correlation prediction method as described in the present invention.

[0020] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0021] The present invention collects data related to terpenoid compounds and their biological activities from multiple public databases and constructs a knowledge graph of the biological activities of terpenoid compounds, which integrates multi-dimensional information such as terpenoid compounds, target proteins, genes, cell lines and diseases, solves the problem of single data source and incomplete coverage in the prior art, and further combines the knowledge graph embedding model to predict related diseases of target terpenoid compounds, providing a reliable data basis for the research of terpenoid compounds.

[0022] The present invention combines molecular physical and chemical characteristics, two-dimensional graph structure characteristics and knowledge graph embedding technology to construct a molecular embedding vector that embeds rich molecular structure and biological information. It further fills in the relationship information in the incomplete knowledge graph through a prediction model based on knowledge graph embedding technology, effectively improving the accuracy of association prediction.

[0023] The present invention can recommend candidate diseases related to terpenoid compounds and predict diseases with higher scores as potential targets, thereby providing important candidate diseases for subsequent experiments or clinical verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 The present invention is a flowchart of a method for predicting the correlation between terpenoid compounds and diseases based on a knowledge graph according to one embodiment.

[0025] Figure 2 The figure is a flow chart of a method for predicting the correlation between terpenoid compounds and diseases according to one embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of a knowledge graph of the biological activity of terpenoid compounds according to one embodiment of the present invention.

[0027] Figure 4 Schematic diagram of the experimental results of testing the anti-melanoma effect of the terpene compound TKC014820 on A375 cells.

[0028] Figure 5 Schematic diagram of the experimental results of testing the anti-melanoma effect of the terpene compound TKC117556 on A375 cells.

[0029] Figure 6This is an architecture diagram of a knowledge graph-based prediction system for terpenoid compounds and diseases according to one embodiment of the present invention. DETAILED DESCRIPTION

[0030] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0031] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "the" used in the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0032] It should be understood that although the terms first, second, third, etc. may be used in the present invention to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0033] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] Example 1

[0035] This embodiment proposes a method for predicting the correlation between terpenoid compounds and diseases based on knowledge graphs, such as Figure 1 , 2 Shown is a flow chart and schematic diagram of the method for predicting the correlation between terpenoid compounds and diseases in this embodiment.

[0036] The method for predicting the correlation between terpenoid compounds and diseases based on the knowledge graph proposed in this embodiment includes the following steps:

[0037] S1. Collect data on terpenoids and their biological activities, including molecular structures, protein targets, gene targets, cell lines and disease association information of terpenoids, and construct a knowledge graph of terpenoid biological activities;

[0038] S2, inputting the molecular descriptors of the terpenoid compounds to be predicted into a convolutional neural network for feature extraction, and generating a first feature vector about the physical and chemical properties of the molecules;

[0039] S3, extracting features of the molecular structure of the terpenoid compound to be predicted through a graph convolutional network to generate a second feature vector about the features of the terpenoid molecular graph;

[0040] S4, embedding the molecular embedding vector obtained by concatenating the first feature vector and the second feature vector into a knowledge graph embedding model to generate a terpenoid compound-disease prediction score matrix;

[0041] S5. Sorting in descending order based on the terpenoid compound-disease prediction scores to generate a recommended list of diseases related to terpenoid compounds.

[0042] This embodiment collects data related to terpenoid compounds and their biological activities from multiple public databases and constructs a knowledge graph of the biological activity of terpenoid compounds, which integrates multi-dimensional information such as terpenoid compounds, target proteins, genes, cell lines and diseases, solves the problem of single data source and incomplete coverage in the prior art, and further combines the knowledge graph embedding model to predict related diseases of target terpenoid compounds, providing a reliable data basis for the research of terpenoid compounds.

[0043] This embodiment further combines molecular physicochemical characteristics, two-dimensional graph structure characteristics and knowledge graph embedding technology to construct a global molecular embedding vector, which embeds rich molecular structure and biological information, making the molecular feature description more comprehensive. The prediction model based on knowledge graph embedding technology fills in the relationship information in the incomplete knowledge graph, effectively improving the accuracy of association prediction. In addition, this embodiment can recommend candidate diseases related to terpenoid compounds, predict diseases with higher scores as potential targets, and provide important candidate diseases for subsequent experiments or clinical verification. This data-driven approach can reduce R&D costs and improve research efficiency.

[0044] In an optional embodiment, when collecting data on terpenoid compounds and their biological activities, data can be collected from public open source databases such as DrugBank, ChEMBL, PubChem, TTD, CTD, GeneCards, Cellosauru, Uniport, TCMBank, TeroKIT, etc. The collected data is further structured and standardized to construct a knowledge graph of terpenoid compound biological activities.

[0045] Exemplarily, substructure information, protein targets, cellular targets and the association data of terpenoid compounds are collected from the TeroKit platform, approved terpenoid small molecule drugs are extracted from the DrugBank database, herbal ingredients classified as terpenes are extracted from the TCMbank database, and cellular targets and their association with rare human diseases are collected from the Cellosaurus database.

[0046] In another example, by integrating the CTD, GeneCards, and TTD databases, using genes as a bridge, protein targets were classified by disease type, and mapping with the TeroKit database was achieved to complete data collection for data analysis related to the interaction between terpenoids and protein targets.

[0047] In an optional embodiment, when constructing the knowledge graph of terpenoid bioactivity in step S1, the following steps are included:

[0048] Preprocessing the collected terpenoid compounds and their biological activity data;

[0049] Integrate the activity associations among five biological entities: terpenoids, protein targets, genes, cell lines, and diseases;

[0050] After filtering the interaction data through a preset activity threshold, the data are stored in the form of triplets to construct a knowledge graph of the bioactivity of terpenoids.

[0051] In this embodiment, the association between terpenoid compounds and targets is integrated, and the targets and diseases are classified using genes as a bridge to obtain the relationship between terpenoid compound entities and disease entities, and a knowledge graph of the bioactivity of terpenoid compounds is constructed in the form of triples (head entity-relationship-tail entity).

[0052] like Figure 3 Shown is the architecture diagram of the terpenoid compound biological activity knowledge graph of this embodiment.

[0053] Exemplarily, the preset activity threshold is set to 10 μM to filter low-confidence interaction data.

[0054] Further optionally, when preprocessing the collected terpenoid compounds and their biological activity data, the following steps are included:

[0055] Standardize the data, including cleaning, identifier unification, and data deduplication.

[0056] This embodiment standardizes the collected data before constructing the knowledge graph, which can adjust the scale of the data so that data with different characteristics have the same distribution characteristics, thereby improving the efficiency and effectiveness of model training.

[0057] In an optional embodiment, in step S2, when the molecular descriptors of the terpenoid compounds to be predicted are input into the convolutional neural network for feature extraction, the following steps are included:

[0058] Calculating molecular descriptors of terpenoid compounds; the molecular descriptors include: Wildman-Crippen distribution coefficient, topological polar surface area, molecular weight, drug-likeness, molar refractive index, number of hydrogen bond acceptors, number of hydrogen bond donors, number of aromatic rings, number of rotatable bonds and carbon atom SP3 hybridization fraction (FractionCSP3);

[0059] All molecular descriptors are spliced ​​to obtain a molecular feature matrix; the molecular feature matrix is ​​input into a convolutional neural network for feature extraction to obtain a first feature vector characterizing the physicochemical properties of terpenoid compounds.

[0060] In this embodiment, molecular descriptors are used to characterize the properties of terpenoid compounds, wherein the molecular descriptors of terpenoid compounds are optionally calculated using the RDKit library.

[0061] Among them, the Wildman-Crippen partition coefficient is used to measure the distribution of compounds in water and lipids and to quantify the hydrophobicity of each atom in the molecule.

[0062] Topological polar surface area (TPSA) represents the surface area of ​​polar atoms in a compound and is often used to evaluate the solubility of the compound, the transportability of the drug within cells, the absorption of the drug in the small intestine in the human body, the permeability of the Caco-2 monolayer and the blood-brain barrier, that is, to predict the ability of molecules to penetrate the cell membrane.

[0063] Molecular weight indicates the molecular weight of a compound, which has a certain influence on the pharmacokinetic properties of the drug.

[0064] Drug-likeness is a comprehensive score calculated based on the physicochemical properties of a drug and is used to assess the potential of a compound as a drug.

[0065] The molar refractive index represents the molecular polarizability, which is related to the molecular volume and density.

[0066] The number of hydrogen bond acceptors is the number of atoms or groups in a compound that are capable of forming hydrogen bonds.

[0067] The number of hydrogen bond donors is the number of atoms or groups in a compound that provide hydrogen bonds.

[0068] The aromatic ring count is the number of aromatic rings in a compound, which affects the stability and binding ability of the molecule.

[0069] The number of rotatable bonds is the number of single bonds in a molecule that can rotate freely, which affects the conformational freedom and flexibility of the molecule.

[0070] Carbon atom sp 3 Hybridization fraction (FractionCSP3) represents the sp 3 The ratio of hybridized carbon atoms is often used to evaluate the complexity of molecular structure.

[0071] In this embodiment, all molecular descriptors are spliced ​​to generate a molecular feature matrix, forming a comprehensive feature representation that combines the physical and chemical characteristics of the molecules, embedding biological information, and making the molecular feature description more comprehensive. The molecular feature matrix is ​​further input into a convolutional neural network, and then sequentially passes through a convolutional layer, a pooling layer, a normalization layer, and a Tanh activation layer to generate a feature vector that characterizes the physical and chemical properties of terpenoid compounds.

[0072] In an optional embodiment, when extracting features of the molecular structure of the terpenoid compound to be predicted through a graph convolutional network in step S3, the following steps are included:

[0073] Generate a corresponding molecular graph based on the molecular structure of terpenoid compounds; wherein, atoms are used as nodes and chemical bonds are used as edges to form a two-dimensional graph structure of terpenoid compound molecules;

[0074] After extracting an atomic adjacency matrix from the molecular graph and generating an atomic feature matrix based on the atomic type, the atomic adjacency matrix and the atomic feature matrix are input into a graph convolutional network to generate a second feature vector representing global molecular features.

[0075] In this embodiment, chemical molecules are represented as a two-dimensional graph structure, and the structural relationship between nodes and edges can be used to capture the structural characteristics of chemical molecules, especially the representation of chemical bonds. Among them, an atomic feature matrix is ​​generated based on atomic types such as C, N, O, S, and the atomic adjacency matrix is ​​extracted from the molecular graph to represent the spatial interconnection of atoms in the molecule, and then input into the graph convolutional network (GCN), and then the global molecular feature vector is generated through the Readout layer, normalization layer and Tanh activation layer in sequence.

[0076] In an optional embodiment, in step S4, the molecular embedding vector obtained by concatenating the first feature vector and the second feature vector is embedded in the knowledge graph model to generate a terpenoid compound-disease prediction score matrix, including the following steps:

[0077] The first feature vector and the second feature vector are concatenated and passed through a fully connected layer to generate a molecule embedding vector;

[0078] Generate disease entity embedding vectors using a random adjacency matrix and generate relationship embedding vectors using a random initialization vector of the same dimension;

[0079] The molecule embedding vector, disease embedding vector and relationship embedding vector are input into the DistMult model, and the score of each triple is calculated to obtain a terpenoid compound-disease prediction score matrix.

[0080] This embodiment combines CNN, GCN and knowledge graph embedding (KGE) to construct a global molecular feature vector and embed rich structural and biological information, making the molecular feature description more comprehensive. Furthermore, the DistMult model is used as the knowledge graph embedding model, which uses a bilinear model to represent the interaction between entities and relationships, aiming to combine the context of knowledge graph and relationship data, represent and infer the relationship between entities, and by filling in the missing relationship information, it helps to improve the current incomplete knowledge graph of terpenoid bioactivity.

[0081] Specifically, the DistMult model utilizes a simple yet effective inner product-based scoring function to capture the second-order correlation between entities, which is expressed as:

[0082] f r (h,t)=h⊙(r⊙t)

[0083] Among them, h, r and t represent the vector representation of the head entity, relation and tail entity respectively, and ⊙ represents the element-by-element product between vectors.

[0084] Further optionally, the scoring function f configured in the DistMult model in this embodiment r It is expressed as:

[0085] f r (t,d)=t T W r d

[0086] Where t represents the molecule embedding vector, d represents the disease entity embedding vector, W r is the weight matrix of the relationship r between the molecule embedding vector and the disease embedding vector.

[0087] In this embodiment, the scoring function f r The credibility of the binary pair (t, d) was evaluated to obtain the terpenoid-disease prediction score.

[0088] Furthermore, according to the scoring function f r The output of is used to sort the prediction scores of all potential terpenoid-disease associations in descending order, and a list of recommended diseases related to terpenoids is generated. The diseases ranked high in the recommended list are considered as potential diseases for the terpenoid.

[0089] The scoring function in this embodiment quantifies the association between each disease and the terpenoid compound based on the characteristic expression information in the terpenoid compound bioactivity knowledge graph constructed in step S1, and can automatically recommend candidate diseases related to terpenoid compounds, predict diseases with higher scores as potential targets, and provide important candidate diseases for subsequent experiments or clinical verification.

[0090] Exemplarily, the structures of two terpenoid compounds TKC014820 and TKC117556 to be tested are input into the model in the form of SMILES sequences. The model includes the knowledge graph of bioactivity of terpenoid compounds proposed in this embodiment, an encoding layer composed of a convolutional neural network, a graph convolutional network, etc., and a decoding layer composed of a knowledge graph embedding model, etc. The model can predict all association scores of the two terpenoid compounds with the 775 disease entities contained in the knowledge graph. According to these scores, they are sorted from high to low, and finally a complete list of disease recommendations is obtained. The diseases ranked high in the list are inferred to be potential related diseases of the two terpenoid compounds.

[0091] The prediction results showed that the prediction scores of melanoma were high, ranking 1 and 2 respectively, indicating the potential association between the two compounds and melanoma. Therefore, the two compounds were used to test their anti-melanoma effects on human malignant melanoma A375 cells.

[0092] The specific operation of the experiment is as follows: A375 cells were seeded in a 96-well plate at a density of 1×10^4 cells per well, fresh medium was added and cultured overnight. After culture, the cells were treated with different concentrations of compounds for 24-48 hours. After removing the supernatant, the cells were labeled with Cell Counting Kit-8 (CCK8) solution according to the manufacturer's instructions. Then incubated at 37°C for 3 hours, and the absorbance was read at 450nm using a microplate reader.

[0093] like Figure 4 , 5 The IC50 values ​​of the two compounds against melanoma at the cellular level are 3.842 nM and 3.27 μM, respectively, indicating that the two terpenoid compounds are indeed significantly associated with melanoma as shown in the prediction results of the model.

[0094] Example 2

[0095] This embodiment applies the terpenoid compound and disease correlation prediction method proposed in Example 1, and proposes a terpenoid compound and disease correlation prediction system based on knowledge graph, such as Figure 6 , which is a schematic diagram of the system for predicting the correlation between terpenoid compounds and diseases according to the present embodiment.

[0096] The terpenoid compound and disease correlation prediction system proposed in this embodiment includes:

[0097] The knowledge graph construction module is used to collect data on terpenoid compounds and their biological activities, including terpenoid molecular structures, protein targets, gene targets, cell lines and disease association information, and to construct a knowledge graph of terpenoid biological activities;

[0098] A first feature extraction module is used to input the molecular descriptors of the terpenoid compounds to be predicted into the convolutional neural network for feature extraction, and generate a first feature vector about the physical and chemical properties of the molecules;

[0099] A second feature extraction module is used to extract features of the molecular structure of the terpenoid compound to be predicted through a graph convolutional network to generate a second feature vector about the features of the terpenoid molecular graph;

[0100] A knowledge graph embedding module, used to embed the molecular embedding vector obtained by concatenating the first feature vector and the second feature vector through a knowledge graph embedding model to generate a terpenoid compound-disease prediction score matrix;

[0101] The output module is used to generate a recommended list of diseases related to terpenoids by sorting them in descending order based on the terpenoid-disease prediction scores.

[0102] It can be understood that the system of this embodiment corresponds to the method of the above-mentioned embodiment 1, and the options in the above-mentioned embodiment 1 are also applicable to this embodiment, so they will not be described repeatedly here.

[0103] Example 3

[0104] This embodiment proposes a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes all or part of the steps of the method for predicting the association between terpenoid compounds and diseases proposed in Example 1.

[0105] Example 4

[0106] This embodiment provides a storage medium having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by a processor, implement all or part of the steps of the method for predicting the correlation between terpenoid compounds and diseases proposed in Embodiment 1.

[0107] Exemplarily, the storage medium includes, but is not limited to, a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0108] Exemplarily, the instructions, programs, code sets or instruction sets may be implemented using conventional programming languages.

[0109] Exemplarily, the processor includes but is not limited to a smart phone, a personal computer, a server, a network device, etc., and is used to execute all or part of the steps of the method for predicting the association between terpenoid compounds and diseases described in Example 1.

[0110] Each embodiment of the present invention is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described above is merely exemplary, in which the modules described as separate components may or may not be physically separated, and the functions of each module can be implemented in the same or one or more software and / or hardware when implementing the scheme of the present invention. It is also possible to select some or all of the modules according to actual needs to achieve the purpose of the scheme of this embodiment.

[0111] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for predicting the correlation between terpenoid compounds and diseases based on knowledge graph, characterized in that: The following steps are involved: Collect data on terpenoids and their biological activities, including molecular structures, protein targets, gene targets, cell lines, and disease association information of terpenoids, and construct a knowledge graph of terpenoid biological activities; The molecular descriptors of the terpenoid compounds to be predicted are input into a convolutional neural network for feature extraction to generate a first feature vector about the physical and chemical properties of the molecules; The molecular structure of the terpenoid compound to be predicted is subjected to feature extraction through a graph convolutional network to generate a second feature vector regarding the features of the terpenoid molecular graph; The molecular embedding vector obtained by concatenating the first feature vector and the second feature vector is embedded in the knowledge graph model to generate a terpenoid compound-disease prediction score matrix; Based on the terpenoid-disease prediction scores, the disease recommendation list related to terpenoids was generated by descending sorting.

2. The method for predicting the correlation between terpenoid compounds and diseases according to claim 1, characterized in that: The method of constructing a knowledge graph of the biological activity of terpenoid compounds comprises the following steps: Preprocessing the collected terpenoid compounds and their biological activity data; Integrate the activity associations among five biological entities: terpenoids, protein targets, genes, cell lines, and diseases; After filtering the interaction data through a preset activity threshold, the data are stored in the form of triplets to construct a knowledge graph of the bioactivity of terpenoids.

3. The method for predicting the correlation between terpenoid compounds and diseases according to claim 2, characterized in that: The preprocessing of the collected terpenoid compounds and their biological activity data comprises the following steps: Standardize the data, including cleaning, identifier unification, and data deduplication.

4. The method for predicting the correlation between terpenoid compounds and diseases according to claim 1, characterized in that: The molecular descriptors of the terpenoid compounds to be predicted are input into a convolutional neural network for feature extraction, comprising the following steps: Calculate molecular descriptors of terpenoid compounds; the molecular descriptors include: Wildman-Crippen distribution coefficient, topological polar surface area, molecular weight, drug-like properties, molar refractive index, number of hydrogen bond acceptors, number of hydrogen bond donors, number of aromatic rings, number of rotatable bonds, and carbon atom sp 3 Hybridization fraction; All molecular descriptors are spliced ​​to obtain a molecular feature matrix; the molecular feature matrix is ​​input into a convolutional neural network for feature extraction to obtain a first feature vector characterizing the physicochemical properties of terpenoid compounds.

5. The method for predicting the correlation between terpenoid compounds and diseases according to claim 1, characterized in that: The molecular structure of the terpenoid compound to be predicted is subjected to feature extraction through a graph convolutional network, comprising the following steps: Generate a corresponding molecular graph based on the molecular structure of terpenoid compounds; wherein, atoms are used as nodes and chemical bonds are used as edges to form a two-dimensional graph structure of terpenoid compound molecules; After extracting an atomic adjacency matrix from the molecular graph and generating an atomic feature matrix based on the atomic type, the atomic adjacency matrix and the atomic feature matrix are input into a graph convolutional network to generate a second feature vector representing global molecular features.

6. The method for predicting the correlation between terpenoid compounds and diseases according to any one of claims 1 to 5, characterized in that: The molecular embedding vector obtained by concatenating the first feature vector and the second feature vector is embedded into a knowledge graph model to generate a terpenoid compound-disease prediction score matrix, comprising the following steps: The first feature vector and the second feature vector are concatenated and passed through a fully connected layer to generate a molecule embedding vector; Generate disease entity embedding vectors using a random adjacency matrix and generate relationship embedding vectors using a random initialization vector of the same dimension; The molecule embedding vector, disease embedding vector and relationship embedding vector are input into the DistMult model, and the score of each triple is calculated to obtain a terpenoid compound-disease prediction score matrix.

7. The method for predicting the correlation between terpenoid compounds and diseases according to claim 6, characterized in that: The DistMult model is configured with a scoring function f r ; Its expression is: f r (t,d)=t T W r d Where t represents the terpene molecule embedding vector, d represents the disease entity embedding vector; W r is the weight matrix of the relationship r between the molecule embedding vector and the disease embedding vector.

8. A terpenoid compound and disease correlation prediction system based on knowledge graph, using the terpenoid compound and disease correlation prediction method according to any one of claims 1 to 7, characterized in that: include: The knowledge graph construction module is used to collect data on terpenoid compounds and their biological activities, including terpenoid molecular structures, protein targets, gene targets, cell lines and disease association information, and to construct a knowledge graph of terpenoid biological activities; A first feature extraction module is used to input the molecular descriptors of the terpenoid compounds to be predicted into the convolutional neural network for feature extraction, and generate a first feature vector about the physical and chemical properties of the molecules; A second feature extraction module is used to extract features of the molecular structure of the terpenoid compound to be predicted through a graph convolutional network to generate a second feature vector about the features of the terpenoid molecular graph; A knowledge graph embedding module, used to embed the molecular embedding vector obtained by concatenating the first feature vector and the second feature vector through a knowledge graph embedding model to generate a terpenoid compound-disease prediction score matrix; The output module is used to generate a recommended list of diseases related to terpenoids by sorting them in descending order based on the terpenoid-disease prediction scores.

9. A device comprising a memory and a processor, wherein the memory stores computer-readable instructions, characterized in that: When the computer-readable instructions are executed by the processor, the processor executes all or part of the steps of the method for predicting the association between terpenoid compounds and diseases according to any one of claims 1 to 7.

10. A storage medium having computer-readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by a processor, all or part of the steps of the method for predicting the correlation between terpenoid compounds and diseases as described in any one of claims 1 to 7 are implemented.

Images