Work

Power transformations improve interpolation of grids for molecular mechanics interaction energies

Year: 2018

Type: article

Abstract: A common strategy for speeding up molecular docking calculations is to precompute nonbonded interaction energies between a receptor molecule and a set of three-dimensional grids. The grids are then interpolated to compute energies for ligand atoms in many different binding poses. Here, I evaluate a smoothing strategy of taking a power transformation of grid point energies and inverse transformation of the result from trilinear interpolation. For molecular docking poses from 85 protein-ligand complexes, this smoothing procedure leads to significant accuracy improvements, including an approximately twofold reduction in the root mean square error at a grid spacing of 0.4 Å and retaining the ability to rank docking poses even at a grid spacing of 0.7 Å. © 2018 Wiley Periodicals, Inc.

Source: Journal of Computational Chemistry

DOI: 10.1002/jcc.25180

Author: David D. L. Minh

Institution: Illinois Institute of Technology

Language: English

Topic: Computational Drug Discovery Methods

Subfield: Computational Theory and Mathematics

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: National Institute of General Medical Sciences

Award: Sound-stage Virtual Screening Based on Implicit Ligand Theory