We develop a new method ProBiS-Dock, an extension of the ProBiS-ligands and ProBiS-CHARMMing approaches (https://probis.nih.gov) for construction of the ProBiS-Dock Database (Figure 1), a web server and repository of small ligand binding sites of different species in the PDB, where each ligand type is exactly determined. Unlike in our previous GenProBiS web server (http://genprobis.insilab.org) in which we determined binding sites on only one protein chain, here we also predict binding sites between multiple protein chains. A protein ligand binding site is defined as the space occupied by the union of all of its predicted ligands obtained from similar binding sites in the PDB using the ProBiS-ligands approach. In standard docking procedures, binding sites are manually curated to prepare them for docking. In contrary, in our newly developed method described here, this binding site curation procedure is performed automatically by selection of the accessory ligands, i.e., the co-crystallized molecules within each PDB file, that should or should not, be present in the specific binding site during docking. These accessory ligands that we classify into compounds, cofactors, metal ions, glycans, and conserved waters may significantly influence drug binding to a protein. This database, freely available for academic users at http://probis-dock-database.insilab.org, is especially useful to address the multi-target drug design, i.e. off-target effects, drug selectivity, etc., and will actively contribute to more efficient development of new drugs.

ProBiS-Dock method and database for efficient development of new drugs

ACS Fall 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.