Method for improving drug-target activity prediction precision by combining machine learning and conformation calculation

Abstract

The invention relates to a method for improving drug-target activity prediction precision in combination with machine learning and conformation calculation, and belongs to the technical field of drug screening, and the method comprises the steps of named entity recognition to obtain a molecular data set in literature, neural network training to output weight parameters, sample clustering and multi-classifier construction. According to the method, a machine learning algorithm and a crystal conformation energy calculation method are combined to improve the precision of virtual screening, so that drug screening is more efficient, the cost is lower, and the accuracy and reliability of results are improved.

Description

technical field [0001] The invention belongs to the technical field of drug screening, and particularly relates to a method for improving the prediction accuracy of drug-target activity by combining machine learning and conformational calculation. Background technique [0002] In the process of drug research, huge labor costs and long research and development cycles have always hindered the advent of new drugs, until the use of computer-based virtual screening technology in the field of medicinal chemistry has accelerated the process of early drug research and development to a certain extent. Traditional molecular docking is a computational-based method that predicts its binding mode and affinity through target characteristics and interactions between drug molecules and targets, for example, using molecular docking software such as Rosetta, Ledock, and AutodockVina. This docking method is still widely used in the field of virtual screening. However, such methods are limited...

AI Technical Summary

Problems solved by technology

The lack of positive samples and the low quality of the data set make it difficult for the model to dig out deep correlations and modes of action, so that the model lacks sufficient generalization

In addition, the ignorance of the target structure information also limits the further breakthrough of the deep learning method to the performance bottleneck, so that it can be practically applied in the field of drug research and development.

Images