How Good Are State-of-the-Art Docking Tools in Predicting Ligand Binding Modes in Protein-Protein Interfaces?

Protein-protein interfaces (PPIs) are an important class of drug-targets. We report on the first large-scale validation study on docking into PPIs. DrugScore-adapted AutoDock3 and Glide showed good success rates with a moderate drop-off, compared to docking to classical targets. An analysis of the binding energetics in a PPI allows identifying those interfaces that are amenable for docking. The results are important for deciding if structure based design approaches can be applied to a particular PPI.