GPCR(G Protein-Coupled Receptor)-drug interaction prediction method based on postprocessing study

Abstract

The invention provides a GPCR(G Protein-Coupled Receptors)-drug interaction prediction method based on postprocessing study. The method comprises the following steps that on the basis of all GPCR-drug information with interaction effects in a training data set, a DAM (drug associated matrix) is built; all of the GPCR-drug information in the training data set is subjected to multi-view-angle feature expression, a training sample set is formed, and then, the training sample set is trained into a GPCR-drug interaction RF (random forest) prediction model by using an RF algorithm; the multi-view-angle feature expression is carried out on each pair of GPCR-drug information with the interaction effect to be predicted, and a set of samples to be predicted is formed; then, the RF prediction model is used for carrying out interaction existence probability prediction, and finally, the prediction probability is output; the postprocessing study is carried out according to the output probability, and finally, the judgment that whether the GPCR-drug information has the interaction effect or not can be directly obtained.

Description

technical field [0001] The invention relates to the field of bioinformatics G protein-coupled receptor-drug interaction, in particular to a method for predicting G protein-coupled receptor-drug interaction based on post-processing learning. Background technique [0002] G protein-coupled receptors (G Protein-Coupled Receptors, GPCRs) is a general term for a large class of membrane protein receptors. What these receptors have in common is that there are seven transmembrane α-helices in their three-dimensional structure, and there are G protein couplings on the C-terminus of the peptide chain and the intracellular loop connecting the fifth and sixth transmembrane helices Binding site for a receptor (guanylate-binding protein). So far, studies have shown that G protein-coupled receptors are only found in eukaryotes and are involved in many cell signal transduction processes. During these processes, G protein-coupled receptors can bind chemicals in the cell's surrounding envir...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the above-mentioned G protein-coupled receptor-drug interaction prediction problem, the lack of correlation information between potential drugs and drugs leads to a large gap between prediction accuracy and poor interpretability, the purpose of the present invention The goal is to propose a post-processing learning-based G protein-coupled receptor-drug interaction prediction method with drug-drug association information and high prediction accuracy

Images