QSAR study of C allosteric binding site of HCV NS5B polymerase inhibitors by support vector machine.

Multiple linear regressions (MLR) and support vector machine (SVM) were used to develop quantitative structure-activity relationship (QSAR) models of novel Hepatitis C virus (HCV) NS5B polymerase inhibitors. Various kinds of molecular descriptors were calculated to represent the molecular structures of compounds, such as chemical, topological, geometrical, and quantum descriptors. Principal component analysis (PCA) was used to select the training set. A variable selection method utilizing a genetic algorithm (GA) was employed to select from the large pool of calculated descriptors, an optimal subset of descriptors which have significant contribution to the overall inhibitory activity. The models were validated using Leave-One-Out (LOO) and Leave-Group-Out (LGO) crossvalidation, and Y-randomization test. Results demonstrated the SVM model offers powerful prediction capabilities.