Implicit ligand theory (ILT) provides a way to calculate noncovalent binding free energies between flexible partners through an exponential average of the binding potential of mean force (BPMF) - the binding free energy between a flexible ligand and rigid receptor [4,10]. Receptor configurations must be drawn from or reweighed to an apo [10] or holo [4] ensemble. Computing binding free energies based on multiple BPMFs has several advantages over simulations with a flexible receptor: the ensemble of receptor configurations can be thoroughly sampled once and recycled for a large chemical library of ligands; accuracy can be progressively increased by computing BPMFs for more receptor snapshots [6]; the task of computing multiple BPMFs is trivially parallel; and BPMFs can be much faster to compute and more scalable than binding free energies to flexible proteins. In the last several years, my group derived ILT [4,10], developed tools to compute protein-ligand BPMFs based on replica exchange [9] and the fast Fourier transform [3], and applied them to the binding of small hydrophobic molecules to T4 lysozyme [6] and in the D3R grand challenge [1]. Along the way, we have contributed methodological advances that have implications beyond protein-ligand binding: to better understanding replica exchange efficiency [8], implemented ergodic Boltzmann sampling based on constrained dynamics [7], improved the interpolation of interaction energy grids [5], and developed a strategy to evaluate ensemble reduction algorithms for docking and BPMF calculations.<br/><br/>

Absolute binding free energies from binding free energies to multiple rigid receptor conformations

2019 Fall National Meeting

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