Prediction method of a G-Protein Coupled Receptor (GPCR) drug and a targeting pathway based on heterogeneous network

Abstract

The invention discloses a prediction method of a G-Protein Coupled Receptor (GPCR) drug and a targeting pathway based on a heterogeneous network and biological application of the method. The method ischaracterized in that a drug-pathway heterogeneous network model is built based on the chemical sub-structure feature of the drug, the SMILES character string information feature of the drug, pathway-related protein sequence features and the phenotypic features of pathway-related diseases, and a drug-pathway interaction (DPI) relationship is predicted by using a deep belief network method. The prediction method of the GPCR drug and the targeting pathway based on the heterogeneous network, provided by the invention, not only is high in prediction precision and robustness, but also can successfully predict the pathway which part of the GPCR drug acts on. The method can be applied to drug-pathway correlation evaluation, clinical patient medication reference and the like.

Description

technical field [0001] The invention relates to the connection between drugs and pathways in bioinformatics, that is, a method for predicting GPCR drugs and targeted pathways based on a heterogeneous network model. The method mainly uses the chemical substructure characteristics of drugs and the SMILES string information characteristics of drugs, And pathway-related disease phenotype similarity and pathway-related protein sequence similarity, so as to construct a drug-pathway heterogeneous network model, and use the deep belief network method to predict potential drug-pathway relationships. Background technique [0002] Identifying drug-pathway interactions (DPI) is the key to drug discovery and drug repositioning. Due to the clear availability of GPCR drugs, if the drug can be used in a new pathway, it can not only reduce the cost of drug development, but also reduce the adverse reactions of the drug. Although various bioassay techniques are currently available to predict ...

AI Technical Summary

Problems solved by technology

The disadvantage of the method is that changing the sign of the impact factor at the same time will not cause the expected impact on the estimated gene expression value, and the data provided are the sensitivity data of the drug and the gene expression data in the cell line cultured by the drug ( References Ma H, Zhao H.iFad:an integrativefactor analysis model for drug-pathway association inference[J].Bioinformatics,2012,28(14):1911-8.MaH,Zhao H

Images