As a deterministic docking program with no random variables, Glide generates results dependent on a small number of factors including the set of simulation parameters, 3D input structures, and receptor grids. In this work we investigate enhanced ligand sampling by taking advantage of ligand 3D input dependence.  The principal factors that govern ligand input dependence are evaluated and different approaches to generating alternative input conformations including minimization,direct torsion adjustment, and canonicalization are developed. Comparison to experimental binding modes demonstrates enhanced ligand sampling via multiple ligand/polypeptide input conformations improves pose accuracy of Glide docking while maintaining its deterministic quality.

Enhanced sampling in ligand and polypeptide docking with Glide

245th National Meeting (2013)

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