Quantum Chemistry for Proteins

Proteins are large biological macromolecules that play a pivotal role in the function of all living things. Because of the large size of proteins (most are at least several hundred to thousands of atoms in size), study of their structure and function has been largely limited to empirical force fields. While these force fields can reproduce many basic structural properties of proteins as observed experimentally by NMR or X-ray crystallography, typical force fields cannot accurately describe bond-rearrangement, polarization, and charge transfer, all of which are key for understanding protein function. We recently investigated whether GPU-accelerated quantum chemistry approaches could provide additional insight into protein structure-function relationships by examining a vast test set of over 55 proteins with a variety of DFT, HF, and force field methods.

We observed that prototypical secondary structure motifs were typically better described by MM methods (as determined by comparison to experimental properties and a "health" score), while proteins with greater regions of disorder were better described by QM methods. This observation is critical to understanding enzyme catalysis in larger proteins because many features of enzyme active sites tend to be disordered or atypical, with key loops and residues out of place from "typical" locations facilitating chemical reactions. As a continuation of this research, we are examining the large-scale quantum effects at enzyme active sites and how that alters our mechanistic predictions of catalytic function.