Assessment of ligand‐binding residue predictions in CASP9

• DOI: 10.1002/prot.23174
• Publication Date: 2011-09-12T11:24:35Z
• Authors: Tobias Schmidt; Jürgen Haas; Tiziano Gallo Cas...; Torsten Schwede
• Keywords: CASP; Matthews correlation coefficient

Abstract Interactions between proteins and their ligands play central roles in many physiological processes. The structural details for most of these interactions, however, have not yet been characterized experientially. Therefore, various computational tools developed to predict the location binding sites amino acid residues interacting with ligands. In this manuscript, we assess performance 33 methods participating ligand‐binding site prediction category CASP9. overall accuracy predictions CASP9 appears rather high (average Matthews correlation coefficient 0.62 10 top performing groups) compared previous experiments more groups performed equally well. However, should be seen context a strong bias test data toward easy template‐based models. Overall, converged similar approach using inference from related homologous structures, which limits applicability difficult de novo targets. Here, present results assessment category, discuss examples successful challenging targets CASP9, finally suggest changes format experiment overcome current limitations assessment. Proteins 2011. © 2011 Wiley‐Liss, Inc.