Docking software performance in protein-glycosaminoglycan systems
Urszula Uciechowska-Kaczmarzyk , Isaure Chauvot de Beauchene , Sergey Samsonov
AbstractWe present a benchmarking study for protein-glycosaminoglycan systems with eight docking programs: Dock, rDock, ClusPro, PLANTS, HADDOCK, Hex, SwissDock and ATTRACT. We used a non-redundant representative dataset of 28 protein-glycosaminoglycan complexes with experimentally available structures, where a glycosaminoglycan ligand was longer than a trimer. Overall, the ligand binding poses could be correctly predicted in many cases by the tested docking programs, however the ranks of the docking poses are often poorly assigned. Our results suggest that Dock program performs best in terms of the pose placement, has the most suitable scoring function, and its performance did not depend on the ligand size. This suggests that the implementation of the electrostatics as well as the shape complementarity procedure in Dock are the most suitable for docking glycosaminoglycan ligands. We also analyzed how free energy patterns of the benchmarking complexes affect the performance of the evaluated docking software.
|Journal series||Journal of Molecular Graphics & Modelling, ISSN 1093-3263, (A 25 pkt)|
|Publication size in sheets||0.5|
|Keywords in English||molecular docking, protein-glycosaminoglycan interactions, modeling glycosaminoglycans, scoring function, binding pose|
|ASJC Classification||; ; ;|
|Score||= 25.0, 24-07-2019, ArticleFromJournal|
|Publication indicators||: 2016 = 0.735; : 2017 = 1.885 (2) - 2017=1.806 (5)|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.