NRWRH for Drug Target Prediction∗

Drug-target interaction prediction is an important problem for the development of novel drugs and human medical improvement. Many supervised and semi-supervised methods are proposed to uncover the relation between drugs and targets. Under the hypothesis that similar drugs target similar target proteins and the framework of Random Walk with Restart, the method of Networkbased Random Walk with Restart on the Heterogeneous network (NRWRH) is proposed to infer potential drug-target relationship. This method integrates three different networks (protein-protein similarity network, drug-drug similarity network, and known drug-target interaction network) into a heterogeneous network by known drug-target interactions and implements the random walk on this heterogeneous network. When applied to four classes of drug-target proteins interaction data including enzymes, ion channels, GPCRs and nuclear receptors, NRWRH significantly improves the previous methods in terms of cross validation.