The performance of HADDOCK (High Ambiguity data Driven DOCKing) on protein-ligand docking cases was measured. More than 80 input (Unbound) structures with conformational change in the binding site ranging from 0.28 Å to more than 2.8 Å were compared to their 7 corresponding native protein-ligand complexes before and after docking. The first goal of this investigation was to determine if a relationship was present between the difference of the Unbound binding site from the Native site and the success of the docking. Docking success (defined as a docked ligand RMSD less than 2 Å) was measured using ligand-specific protocols for each Unbound case. The second goal of this investigation was to determine if a large conformational difference in the binding site of an Unbound structure has a greater effect on docking success than a large conformational difference farther from the binding site. The relationship between the location of conformational differences and the success of docking will be presented through a comparison of the RMSDs of shells of residues from regions at varying distances from the binding site of the protein structure to the success of the docking for that structure. The final goal was to determine how extensive the conformational changes to an Unbound must be for docking success and if there was a relationship between the Unbound site RMSD and the amount of conformational change observed during docking. With this information, it will be possible to determine if HADDOCK is generally applicable to most protein-ligand docking cases. By knowing which types of Unbound differences prevent successful docking, we can predict the success of typical docking scenarios where the Unbound structure differs from what the bound complex will be.

Protein-ligand docking with HADDOCK: The effects of input protein conformational differences on the success of docking

253rd National Meeting (2017)

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