An overview of data‐driven HADDOCK strategies in CAPRI rounds 38‐45

• DOI: 10.1002/prot.25869
• Publication Date: 2019-12-30T12:23:06Z
• Authors: Panagiotis I. Koukos; Jorge Roel‐Touris; Francesco Ambrosetti; Cunliang Geng; Jörg Schaarschmidt; Mikael E. Trellet; Adrien S. J. Melq...; Li C. Xue; Rodrigo V. Honorato; Irina Moreira; Zeynep Kurkcuoglu; Anna Vangone; Alexandre M. J. J...
• Keywords: Robustness; Docking (animal); Template

Abstract Our information‐driven docking approach HADDOCK has demonstrated a sustained performance since the start of its participation to CAPRI. This is due, in part, ability integrate data into modeling process, and robustness scoring function. We participated CAPRI both as server manual predictors. In rounds 38‐45, we have used various strategies depending on available information. These ranged from imposing restraints few residues identified literature being important for interaction, binding pockets homologous complexes or template‐based refinement/CA‐CA restraint‐guided templates. When relevant, symmetry were limit conformational sampling. also tested large decamer target new implementation MARTINI coarse‐grained force field HADDOCK. Overall, obtained acceptable better predictions 13 11 submissions, respectively, out 22 interfaces. (acceptable higher‐quality models when considering top 10) was (59%) than (50%) one, which typically experiment with combinations protocols sources. Again, our simple function based linear combination intermolecular van der Waals electrostatic energies an empirical desolvation term good 63% success rate across all An analysis model quality indicates that, while are consistently performing well generating models, there room improvement generating/identifying higher models.