New methods for treating electrostatics and adaptive partitioning in QM/MM simulations

University Of Wisconsin-Madison, Madison WI

Investigators

Abstract

Qiang Cui of the University of Wisconsin, Madison is supported by an award from the Theory, Models and Computational Methods program to develop accurate but efficient quantum mechanics/molecular mechanics (QM/MM) methods for use in biological and complex chemical systems. The PI and his research group are pursuing effective electrostatic treatments for QM/MM simulations that employ finite sphere boundary conditions, thus helping to focus configurational sampling to the active site and making ab initio QM/MM simulations cost effective. They are developing a microscopic/mesoscopic model in which atoms within a spherical region are treated atomistically while the rest of the protein is represented by polarizable dipoles at the residue level; the bulk solvent is a Gaussian field, which is equivalent to a dielectric continuum in the static limit. One long term goal is to apply this new approach to identify molecular factors that dictate transition metal binding specificity in systems like metal chaperones and metal specific transcription factors. The methods are also being applied to chemical processes that explicitly involve many water molecules, either in bulk solution, the air/water interface or the interior of biomolecules. The novel computational methods that the PI and his research group develop may be used to study a broad range of important problems in chemistry, biology, medicine and environmental science that are currently difficult or impossible to study. These theoretical/computational studies have the potential to reveal surprising mechanisms at the microscopic level that help explain macroscopic features of the systems. These methods are implemented in the popular simulation package CHARMM and are widely used by other scientific groups.