An Efficient Implementation of the Nwat-MMGBSA Method to Rescore Docking Results in Medium-Throughput Virtual Screenings

The Nwat-MMGBSA method, whose theory has been described in Maffucci & Contini, JCTC 2013, 9, 2706, is based on the inclusion as part of the receptor of a given number of water molecules (Nwat) which are the closest to a residue (generally the ligand) or to a selection of residues (the contact interface) in each frame of the MD simulation. The method was shown to improve the correlation between predicted and experimental binding energy in both ligand-receptor and protein-protein complexes (Maffucci & Contini, JCIM 2016, 56, 1692). Here, we report on the optimization of the Nwat-MMGBSA protocol for its use to rescore docking results. We also report an automatic workflow, based on three independent scripts (which can be concatenated in a fully automated procedure) to easily employ Nwat-MMGBSA rescoring in virtual screening application. The protocol has been tuned using three different examples, and then tested in two retrospective virtual screening examples. In each example, the Nwat-MMGBSA method has been compared with the standard MMGBSA approach (Nwat=0). A link to download the scripts, working examples and tutorials is also provided.