Application of asymmetric statistical potentials to antibody–protein docking

• DOI: 10.1093/bioinformatics/bts493
• Publication Date: 2012-10-10T10:43:38Z
• Authors: Ryan Brenke; David R. Hall; Gwo-Yu Chuang; Stephen R. Comeau; Tanggis Bohnuud; Dmitri Beglov; Ora Schueler-Furman; Sandor Vajda; Dima Kozakov
• Keywords: Paratope; Docking (animal); Macromolecular docking

Abstract Motivation: An effective docking algorithm for antibody–protein antigen complex prediction is an important first step toward design of biologics and vaccines. We have recently developed a new class knowledge-based interaction potentials called Decoys as the Reference State (DARS) incorporated DARS into program PIPER based on fast Fourier transform correlation approach. Although was best performer in latest rounds CAPRI protein experiment, it much less accurate pairs than other types complexes, spite incorporating sequence-based information location paratope. Analysis complexes has revealed inherent asymmetry within these interfaces. Specifically, phenylalanine, tryptophan tyrosine residues highly populate paratope antibody but not epitope antigen. Results: Since this cannot be adequately modeled using symmetric pairwise potential, we removed usual assumption symmetry. Interaction statistics were extracted from under that particular atom different same protein. The use potential significantly improves performance even without any sequence note asymmetric captures effects multi-body interactions to environment interface. Availability: method implemented ClusPro server, available at http://cluspro.bu.edu. Contact: [e-mail] or [e-mail] Supplementary information: data are Bioinformatics online.