You are visitor No: 97871



Prediction of Chemical-Protein Interactions with an Improved Network-Based Inference Method

Chemical-protein interaction (CPI) is the central theme target identification and drug discovery. Large scale testing CPI is a big challenge for experimentally alone. Computational method is appealing due to low cost and high predictive accuracy. In this study, we further developed the node and edge weighted network-based inference (NBI) methods for CPI prediction. Two comprehensive CPI bipartite networks about 17,100 CPI pairs among 4,741 compounds and 97 G protein-coupled receptors (GPCRs), and 13,600 CPI pairs among 2,827 compounds and 206 kinases extracted from the ChEMBL were constructed. The area under the receiver operating characteristic curve was about 0.98 and 0.83 for the test sets and the external validation sets, respectively. The weak interactions hypothesis in CPI network was first proposed by the edges-weighted NBI method. To showcase our method, several candidate targets were predicted for five approved drugs, namely imatinib, dasatinib, sertindole, olanzapine and ziprasidone. The molecular hypotheses and experimental evidences for these predictions were further provided. These results confirmed that our methods have potential values to find molecular basis of drug polypharmacology and drug repositioning.

Experimental Data sets

Predicted Chemical-Protein Interactions database



Feixiong Cheng, Yadi Zhou, Weihua Li, Guixia Liu and Yun Tang. Prediction of Chemical-Protein Interactions Network with Weighted Network-Based Inference Method. PLoS One, 2012, 7(7):e41064.