Compound-target protein binding prediction method

Abstract

The invention discloses a compound-target protein binding prediction method, and relates to the field of medicine, and the method comprises the following steps: (1) obtaining compound-target protein binding / non-binding data; (2) extracting a compound-target protein binding / non-binding data set for training, testing and verification, respectively constructing and training compound-target protein binding / non-binding deep learning models, extracting compound-target protein binding / non-binding features, and obtaining a final compound-target protein binding / non-binding deep learning model; (3) conducting predicting through the final compound-target protein binding / non-binding deep learning model in the step (2) to obtain a compound-target protein binding / non-binding relationship; and (4) allowing the following equation to be met: a simplified compound-target protein binding relation = binding relation - non-binding relation. The compound-target protein binding prediction method provided by the invention is low in false positive rate and high in accuracy.

Description

technical field [0001] The invention belongs to the field of pharmaceutical research and development, and relates to a compound-target protein binding prediction method. Background technique [0002] Using the existing mainstream deep learning models (fully connected network model, convolutional neural network model, recurrent neural network model, encoder-decoder network and deep belief network), extract the characteristics of the compound-target protein binding relationship, so that Predicting the new compound-target protein binding relationship has important practical significance for the discovery / development of new drugs and the study of the mechanism of action of traditional Chinese medicine. However, although these models can achieve a high accuracy rate (>90%), the high false positive rate (<1%) prevents the further application of deep learning models in this field. How to reduce the false positive rate of compound-target protein binding relationship predictio...

AI Technical Summary

Problems solved by technology

[0003] Existing techniques exploit compound-target protein binding relationships (see appendix figure 1 ) cheminformatics software, all manual, many parameters need to be adjusted, the success rate is low, and there is a problem of high false positive

Deng, N et al. (J.Phys.Chem.B2015,119,976-988.) and Nataraj S et al. (Biophys Rev (2017) 9:91–102.DOI 10.1007 / s12551-016-0247-1) from the compound -Target protein binding free energy calculation (free energy calculation) and compound-target protein binding software review discusses the existence of many parameters in compound-target protein binding prediction, low success rate, and high false positives

Images