Prediction of binding sites in HCV protein complexes using a support vector machine

Abstract

Hepatitis C virus (HCV) infection is a major cause of liver disease and a dangerous threat to public health. Hence, the problem of interactions between HCV and human proteins has received much attention. In the present study, we propose a support vector machine (SVM) model for predicting the binding residues in HCV protein complexes. The SVM model achieved an average sensitivity of 76.06% and specificity of 75.94% for 18 non-redundant HCV protein complexes. This approach can efficiently search potential protein-binding sites in proteins and a wide range of protein-protein interaction sites. © 2008 Springer-Verlag Berlin Heidelberg.