Drug-target affinity prediction method based on consistent expression of heterogeneous data
Boyuan Liu · 2022
The first step in drug discovery is to find drug molecule moieties with medicinal activity against specific targets, so it is crucial to investigate the interaction between drug target proteins and chemical small molecules. However, traditional experimental methods for discovering potential drug small molecules are labor-intensive and time-consuming, and there is currently a lot of interest in building computational models to screen drug small molecules by using drug molecule-related databases. In this paper, we propose a method for predicting drug-target binding affinity using deep learning models. The method uses a modified GRU and GNN to extract features from the drug-target protein sequence and the drug-molecule map, respectively, to obtain their feature vectors, and finally the combined vectors are used as vector representations of drug-target molecule pairs into a fully-connected network to predict drug-target binding affinity. The proposed model demonstrates its effectiveness and accuracy in the task of predicting drug-target binding affinity on DAVIS and KIBA datasets.