Computational study of the binding mode of epidermal growth factor receptor kinase inhibitors.

Epidermal growth factor receptor kinase is relative to the progression of various types of cancers. In order to design anticancer drug, docking and support vector machines were used to guide CoMFA and CoMSIA for constructing optimal 3D-QSAR model. Additional descriptors, log P and HOMO, combined with several fields of CoMFA and CoMSIA, were introduced to construct models for the inhibitor of epidermal growth factor receptor kinase. The results show that the inclusion of log P and HOMO energy is meaningful for 3D-QSAR model. The validation of these models was testified by some structurally diverse compounds, which were not included in the CoMFA and CoMSIA models. The docking study and molecular dynamic simulation permit us to insight into the binding mode between ligand and EGFR kinase, and provide important information for structure-based drug design. The proposed approach can also be extended to other QSAR investigations.