Method of predicting interaction between chemical compounds and proteins based on random forest

Abstract

The invention discloses a method of predicting interaction between chemical compounds and proteins based on random forest. The method includes: collecting information of target proteins which tend to interact with drug compounds to establish a target base; collecting the drug compounds for establishment of a training set and information of interaction relation between the drug compounds and the target proteins to establish a compound base; establishing the training set according to the information of the compound base and the target base; training by modified random forest algorithm on the basis of the training set to establish a predicting model; collecting the compounds for prediction and the information of the target proteins obtained in step (A) to establish a test set; predicting the test set on the basis of the predicting model; and (H), judging whether interaction between the compounds to be predicted and the target proteins exists or not according to the prediction result. Accuracy of predicting interaction between the compounds and the proteins by the method can be improved.

Description

technical field [0001] The invention relates to the field of computer-aided drug design, in particular to a method for predicting the interaction between compounds and proteins based on random forest algorithm. Background technique [0002] In the past ten years, although the investment in drug research and development has increased worldwide, its output—the number of drugs approved by the FDA has shown a downward trend year by year (C.R.Chong&D.J.Sullivan, Nature , 2007.448: p.645-646.). More and more scholars believe that the traditional "single drug, single target" drug development model is the main reason for this result. [0003] The goal of traditional drug R&D is to discover drugs targeting a single target with high selectivity and high safety. However, people have gradually discovered that the relationship between drugs and targets is a many-to-many relationship, that is, one drug often acts on multiple different targets, and one target often interacts with multipl...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to propose a method based on the random forest algorithm that can be used to discover the interaction between compounds and proteins in order to improve the efficiency of compound and protein interactions more effectively. The accuracy of the interaction prediction

Images