Prediction drug-target interaction networks based on semi-supervised learning method
Quan Gu, Yongsheng Ding, Tongliang Zhang, Tao Han · 2016
Predicting interactions between drugs and target proteins is a crucial step to decipher many biological processes, and plays a critical role in drug discovery. In this work, we present an Improved Laplacian Regularized Least Square Method (ILRLS) for drug-target interaction prediction. We predict unknown drug-target interactions from chemical structure information, genomic sequence information simultaneously and drug-protein interaction network space. We obtain the better achievement from enzymes, ion channels, nuclear receptors and GPCRs interaction networks. The result indicates that the method could play a complementary role to the existing prediction methods.