DIORS: Enhancing drug-target interaction prediction via structure and signature integrated-driven approach and discovering potential targeted molecules

Abstract

Drug-target interaction prediction is critical for drug development. Through the integration of structural and transcriptional signature information, molecules both binding to the target and producing therapeutic activities could be found out to improve targeted drug prediction. Therefore, the approaches that integrate the two types of data are worth exploring. Here, we present an integrated method named Data Integration Oriented Repurposing Strategy (DIORS) combining molecular docking and gene-signature matching to enhance the prediction of protein-targeted drugs. The StandardScaler algorithm was selected after evaluation of five algorithms and was used in DIORS. Surface Plasmon Resonance (SPR) was used to verify the molecular affinities and cell-based assays were used to verify the activities of DIORS predicted molecules. In Piezo1-targeted molecule prediction, among the top ten predicted molecules by DIORS, four of them, namely gefitinib, rifaximin, bosutinib and vandetanib, exhibited binding affinities. In the prediction of TLR4/MD2-targeted anti-inflammatory molecules, among the top ten predicted molecules, three of them, namely enoxolone, dabrafenib and ponatinib, exhibit both high binding affinities and anti-inflammatory activities. The results demonstrated that DIORS can serve as a better approach with high performance to predict and find new targeted drugs by combining structural and signature information.