Modeling method and device of compound toxicity prediction model and application of compound toxicity prediction model

Abstract

The invention provides a modeling method of a compound toxicity prediction model. The modeling method at least comprises the following steps of: S101, providing toxicity classification labels of candidate modeling compounds; S102, providing a molecular descriptor of each candidate modeling compound; S103, providing a target protein descriptor of each candidate modeling compound; S104, providing aquantitative high-throughput screening analysis descriptor of each candidate modeling compound, wherein the quantitative high-throughput screening analysis descriptor is a PubChem activity score of aspecified amount of high-throughput screening; and S105, constructing and training a compound toxicity prediction model. According to the method, physicochemical properties, biological activity and target protein action properties of drug candidate compounds can be fully utilized, and a drug toxicity prediction system is constructed by utilizing statistical modeling advantages of a machine learning algorithm based on ensemble learning, so that the model has interpretability and prediction performance, and has the better physicochemical and biological significance and research value.

Description

technical field [0001] The invention relates to the fields of chemical informatics and bioinformatics, in particular to a compound toxicity prediction model modeling method, device and application thereof. Background technique [0002] The modern drug development process is a process of exploring compounds that can interact with specific therapeutic targets and have good absorption, distribution, metabolism and excretion properties. According to statistics, about 30% of new drug development failures are caused by safety issues. In order to effectively improve the rate and success rate of drug development, it is necessary to evaluate the toxicity of compounds in the early stage of drug development and exclude the more toxic compounds as soon as possible. Traditional drug toxicity prediction methods mainly use toxicological experimental methods based on animal experiments to predict drug toxicity. Due to the need to verify the actual application effect of the drug on live an...

AI Technical Summary

Problems solved by technology

The shortcomings of traditional drug toxicity prediction methods, as well as the characteristics of current social development, have posed challenges for drug developers

Images