CN-121983114-A - Traditional Chinese medicine intelligent matching system based on network targeting comprehensive index and screening method thereof

CN121983114ACN 121983114 ACN121983114 ACN 121983114ACN-121983114-A

Abstract

A traditional Chinese medicine intelligent screening method based on network targeting comprehensive indexes belongs to the technical field of traditional Chinese medicine intelligent screening. The method comprises the steps of constructing a target set A and a protein interaction network database corresponding to each traditional Chinese medicine, constructing a disease core gene set B based on a protein interaction network, calculating a network targeting comprehensive index based on the target set A and the disease core gene set B, carrying out ascending sort after calculating the network targeting comprehensive index of each traditional Chinese medicine, wherein the smaller the network targeting comprehensive index is, the stronger the network targeting relevance between the traditional Chinese medicine and the disease is, matching the traditional Chinese medicine according to the obtained ascending sort, and generating a structured and interpretable final recommendation report. The invention improves the identification accuracy of the related genes of the diseases, enhances the biological interpretation of the results, realizes the standardized processing of integrating the multisource data into intelligent recommendation, and meets the urgent requirements of high-efficiency and accurate screening in the modern research of traditional Chinese medicines.

Inventors

ZHOU YINGYU
ZHANG CUICAN
LU JIA
LIANG MEIHUA
ZHOU YANG
XIAO DAN
LU WEIHONG

Assignees

哈尔滨工业大学
哈工大郑州研究院

Dates

Publication Date: 20260505
Application Date: 20251226

Claims (5)

1. A traditional Chinese medicine intelligent screening method based on network targeting comprehensive indexes is characterized by comprising the following steps: s1, constructing a target set A and a protein interaction network database corresponding to each traditional Chinese medicine; S2, constructing a disease core gene set B based on the protein interaction network constructed in the S1; s3, calculating a network targeting comprehensive index based on the target set A and the disease core gene set B ; S4, calculating network targeting comprehensive index of each traditional Chinese medicine Then, ascending order is carried out, and network targeting comprehensive index is carried out The smaller the size, the stronger the network targeting association of the traditional Chinese medicine and the disease is; S5, matching the traditional Chinese medicines according to the ascending order obtained in the step S4, and generating a structured and interpretable final recommendation report.
2. The intelligent screening method of traditional Chinese medicine based on network targeting comprehensive index according to claim 1, wherein S1 comprises the following steps: S101, acquiring traditional Chinese medicine component information from a plurality of traditional Chinese medicine databases, and acquiring a union set of traditional Chinese medicine components in each database to form a component data set of the traditional Chinese medicine; S102, mapping traditional Chinese medicine components in the component data set to verified protein targets through compound target data verified by an STITCH database to form a target data set, and de-duplicating targets corresponding to all components contained in each traditional Chinese medicine to be used as a target set A of the traditional Chinese medicine; and S103, constructing or loading a human protein interaction network based on experimental verification data to form a protein interaction network database.
3. The intelligent screening method of traditional Chinese medicine based on network targeting comprehensive index according to claim 2, wherein S2 comprises the following steps: S201, acquiring a disease-related histology data set, and performing quality control, standardization and normalization pretreatment to obtain pretreated gene expression data; S202, screening expression difference genes of a disease group and a control group according to a preset differential expression threshold value based on the preprocessed gene expression data to form an initial seed gene set S; S203, mapping the initial seed gene set S to a protein interaction network to form a network node, and constructing an initial probability vector If the network node belongs to the initial seed gene set S, the value is assigned as , The number of genes in the initial seed gene set S is set to be 0; S204, executing a restarting random walk algorithm to iterate and converge until the probability distribution converges to obtain a steady probability distribution The iterative formula is: wherein: Is the first Probability distribution vectors of steps; A column random adjacency matrix for the network; The method is a preset restarting probability; S205, according to the steady state probability distribution And (3) ordering all the network nodes in a descending order, and selecting the network nodes with the top K rank to form a disease core gene set B.
4. The intelligent screening method of traditional Chinese medicine based on network targeting comprehensive index according to claim 3, wherein S3 comprises the following steps: S301, calculating targets of target set A Genes associated with disease core gene set B Average shortest path distance in protein interaction networks : Wherein: for all targets with communication paths And genes Number of pairs; As target spot And genes Shortest path length in protein interaction network; S302, calculating average shortest path distance in target set A : Wherein: The number of targets is the target set A; Is the first in the target set A And (b) Target points; set A as target And (b) Shortest path length of each target point; s303, calculating the average shortest path distance inside the disease core gene set B : Wherein: the number of targets is target set B; is the first in the target set B And (b) Target points; is the first in the target set B And (b) Shortest path length of each target point; S304, calculating network targeting comprehensive index : 。
5. The intelligent screening method for traditional Chinese medicine based on network targeting complex index according to any one of claims 1 to 4, wherein the method is completed by a traditional Chinese medicine intelligent matching system based on network targeting complex index, and the intelligent traditional Chinese medicine matching system based on network targeting complex index comprises: The database construction module is used for executing S1, and constructing a target set A and a protein interaction network database corresponding to each traditional Chinese medicine; The disease core gene analysis module is used for executing S2 and constructing a disease core gene set B; the topology analysis calculation module is used for executing S3 and calculating network targeting comprehensive indexes ; The traditional Chinese medicine matching and sorting module is used for executing S4 and is based on network targeting comprehensive indexes Ascending order sorting is carried out on the traditional Chinese medicines; And the report generation module is used for S5, and outputting a recommendation result.

Description

Traditional Chinese medicine intelligent matching system based on network targeting comprehensive index and screening method thereof Technical Field The invention relates to a traditional Chinese medicine intelligent screening method based on a network targeting comprehensive index, and belongs to the technical field of traditional Chinese medicine intelligent screening. Background With the breakthrough development of systems biology and network pharmacology, drug development models are being transformed from traditional single-target research to multi-target, networked research paradigms. As a core carrier for functions such as in vivo signal transduction and metabolic regulation, a Protein-Protein interaction network (PPIN) has become a key analysis tool for analyzing the mechanism of action of drugs and the occurrence and development rules of diseases. However, a systematic analysis framework capable of automatically and intelligently integrating protein interaction network data, drug target information and disease gene data is not formed in the prior art, and the deep mining capability of a network topology structure is especially lacked, so that key links such as network distance analysis, drug recommendation and the like still depend on manual data processing and fixed format input, and the technical defects of low analysis efficiency, insufficient method flexibility, weak result interpretation and the like exist. The traditional Chinese medicine is taken as an important component of the traditional medical system, and the overall regulation characteristics of multiple components, multiple targets and multiple paths of the traditional Chinese medicine have natural compliance with the pathological characteristics of multi-factor pathopoiesia of modern diseases. But is limited by the traditional single-target drug research paradigm, and the analysis of the complex action mechanism of traditional Chinese medicine faces technical bottlenecks for a long time. In recent years, a network pharmacological method based on a protein interaction network provides a new technical path for researching the action mechanism of traditional Chinese medicines, and related tools and databases are widely applied in links such as target spot identification, path analysis and the like. However, under the drive of the actual demands of modern research of traditional Chinese medicines, the prior art system still lacks a systematic analysis framework capable of automatically fusing multidimensional network pharmacological indexes (such as network topology parameters, target correlation degree, path coverage rate and the like) and generating comprehensive quantitative scoring indexes, and is difficult to realize multi-level intelligent matching and medicine optimization from target points-paths-networks, so that a more efficient and accurate intelligent screening method of traditional Chinese medicines is needed to be constructed through technical integration and intelligent upgrading. Disclosure of Invention In order to solve the problems in the background technology, the invention provides a network targeting comprehensive index-based intelligent traditional Chinese medicine screening method. The intelligent screening method of the traditional Chinese medicine based on the network targeting comprehensive index comprises the following steps: s1, constructing a target set A and a protein interaction network database corresponding to each traditional Chinese medicine; S101, acquiring traditional Chinese medicine component information from a plurality of traditional Chinese medicine databases, and acquiring a union set of traditional Chinese medicine components in each database to form a component data set of the traditional Chinese medicine; S102, mapping traditional Chinese medicine components in the component data set to verified protein targets through compound target data verified by an STITCH database to form a target data set, and de-duplicating targets corresponding to all components contained in each traditional Chinese medicine to be used as a target set A of the traditional Chinese medicine; and S103, constructing or loading a human protein interaction network based on experimental verification data to form a protein interaction network database. S2, constructing a disease core gene set B based on the protein interaction network constructed in the S1; S201, acquiring a disease-related histology data set, and performing quality control, standardization and normalization pretreatment to obtain pretreated gene expression data; S202, screening expression difference genes of a disease group and a control group according to a preset differential expression threshold value based on the preprocessed gene expression data to form an initial seed gene set S; S203, mapping the initial seed gene set S to a protein interaction network to form a network node, and constructing an initial probability vector If the network node belongs t