Method for predicting direct targets and action components of traditional Chinese medicine compound

Abstract

The invention discloses a method for predicting direct targets and action components of a traditional Chinese medicine compound, and the method comprises the following steps: firstly, mining a gene expression database, and establishing a target knockout / overexpression characteristic gene database; collecting active ingredients possibly existing in the compound in a database, performing modeling, and extracting molecular electronic fingerprint features; based on the characteristic gene database, performing genome enrichment analysis on the omics data, and screening possible direct targets of the compound; for the predicted target points, screening small molecules with verified activity based on a public database, and recording activity properties and activity data of the small molecules; establishing a prediction model for the active data based on a regression machine learning model, and performing model prediction on the active property based on a classifier machine learning model; andpredicting the prescription active ingredients collected in the prediction based on the established prediction model. The effective components in the traditional Chinese medicine compound are successfully predicted through a molecular docking and machine learning combined method.

Description

technical field [0001] The invention relates to the field of traditional Chinese medicine, in particular to a method for predicting direct targets and active components of a compound traditional Chinese medicine. Background technique [0002] An important task of the modernization of traditional Chinese medicine is to use modern science and technology to inherit and develop the discipline of traditional Chinese medicine, so as to seek new vitality. With the continuous development of science and technology, the research on compound Chinese medicine has also made great progress. A batch of active ingredients represented by artemisinin have been discovered, injecting new blood into the ancient discipline of Chinese medicine. However, in the past 40 years, the vast majority of compound research has been to extract effective monomers in compound prescriptions, and use purified effective monomers to replace traditional medicines such as decoctions and powders. Outstanding Contrib...

AI Technical Summary

Problems solved by technology

But there are still two important problems: 1. The current high-throughput screening methods are often aimed at molecular biological analysis and targeted drug development, and there is no strategy tailored to the characteristics of traditional Chinese medicine. Therefore, how to establish a TCM-based The overall strategy of compound characteristics, step by step analysis and prediction of the material basis of the composition is a key difficulty; 2. High-throughput data is often accompanied by massive information. For example, the current one-time sequencing depth can reach more than 10G. In addition, a large number of databases include expression profiles, How to apply big data processing methods to analyze information such as transcript profiles, chemical components, and drug biological activities is also an important issue

[0008] a.Docking technology requires precise protein crystal model and active site location, the cost of obtaining such information is relatively high, once it is not reported, the implementation of Docking becomes very difficult;

[0009] b. Docking can only predict small molecules that directly interact with proteins, and cannot be used to regulate activity through transcription or other indirect modes of action;

[0010] c. The mainstream of Docking is to evaluate affinity based on free energy changes and spatial structure, and cannot directly judge the mode of action (ie activity) of small molecules on proteins

However, the prescription information of traditional Chinese medicine is often not in the mode of chimeric docking. Simple prediction of active ingredients in the prescription based on Docking technology will lose a lot of information, which is not conducive to the true and accurate prediction of the material basis of the compound prescription.

Images