Free energy calculation is a powerful tool in structure-based drug design, pKa predictions, and mechanistic studies of the versatile functions of biomolecules. Different biomolecules show different levels of structural flexibility, which plays an important role in many biological processes, including but not limited to ligand binding, and drug partitioning across the cell membrane. Poor description of such conformational flexibility may lead to low precision and/or low accuracy in free energy calculations. In this work, to better describe potential conformational rearrangement for a given system, we combined an enhanced sampling method, replica exchange with solute tempering (REST), with two types of free energy methods, multistate Bennett acceptance ratio (MBAR) and multisite  dynamics (MSD), to predict pKa of titratable residues in a series of SNase variants, pKa of a small molecule bound with dihydroorotate dehydrogenases (DHOD), and relative binding free energies of multiple small molecules to a common receptor T4 lysozyme (T4L/L99A). Our results suggest that increased structural flexibility could be observed when REST is utilized, which leads to faster convergence of free energy calculations and better agreement with experimental measurements. Certain limitations of this approach have also been identified, such as sampling of large-scale conformational rearrangement like reversible folding/unfolding of proteins. For these complex cases, more advanced sampling methods, such as machine learning based approaches, may be required.

Improving free energy calculations using replica exchange with solute tempering

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

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