Abstract
Predicting the binding structure of bio-complex is essential for understanding its properties, functions, and mechanisms, but is rather difficult due to the huge sampling space involved. A new computational protocol, MDO, for finding the ligand binding structure is proposed. MDO consists of global sampling via MD simulation and clustering of the receptor configurations, local sampling via molecular docking and clustering of the ligand conformations, and binding structure optimization by the ONIOM (QM/QM) method. MDO is tested on 15 protein-ligand complexes with known accurate structures. The success rate of MDO predictions, with RMSD < 2 Å, is found to be 67%, substantially higher than the 40% success rate of conventional methods. The MDO success rate can be increased to 83% if the ONIOM calculations are applied only for the starting poses with ligands inside the binding cavities. The MDO protocol is a promising tool for the structure based drug design.