[Computer drug design based on analysis of a target macromolecule structure. I. Search and description of a ligand binding site in a target molecule]

Abstract

The applicability of molecular docking method for finding of ligand binding site in a target protein was tested. The basic principle of tested approach consists in generation of hypotheses of protein-ligand complexes by molecular docking of small ligand to all surface of protein. The subsequent scoring of these hypotheses utilizes the values of contact surfaces and complex formation energy. The docking procedure was executed using the original software DockSearch running on PC Pentium or SGI computers. A set of hypotheses of probable ligand positions on the protein surface was created and evaluated with the help of DockSearch. Energy minimization of molecular complexes was done using Sybyl 6.5 (Tripos Inc.) running on SGI server Origin 200. The final set of the best hypotheses of complexes were selected by the values of contact surfaces and complex formation energy. The applicability and the limits of this approach was tested using known 3D structures of different proteins in free state and in complexes with ligands. For most target proteins the method allows to find the ligand binding site correctly. The accuracy of description of ligand binding site is adequate for subsequent searching of lead compounds by database mining. The reasons of some negative results obtained in testing of this approach are also discussed.