PocketPicker: analysis of ligand binding-sites with shape descriptors

摘要: Identification and evaluation of surface binding-pockets occluded cavities are initial steps in protein structure-based drug design. Characterizing the active site's shape as well distribution surrounding residues plays an important role for a variety applications such automated ligand docking or situ modeling. Comparing similarity binding site geometries related proteins provides further insights into mechanisms binding. We present PocketPicker, grid-based technique prediction pockets that specifies potential binding-site with regard to its buriedness. The method was applied representative set protein-ligand complexes their corresponding apo-protein structures evaluate quality predictions. performance pocket detection routine compared results achieved existing methods CAST, LIGSITE, LIGSITEcs, PASS SURFNET. Success rates PocketPicker were comparable those LIGSITEcs outperformed other tools. introduce descriptor translates arrangement grid points delineating detected correlation vector. show this is suited comparative analyses similar geometry by examining induced-fit phenomena aldose reductase. This new uses information derived from calculations buriedness binding-sites. useful tool identification binding-pockets. It produces convenient representation shapes including intuitive description accessibility. shape-descriptor classification can be used additional complementing elaborate manual inspections.