CA
Virtual Dissertation Defense
Insights into the Modeling of Crystallographic Structure and Disorder From Molecular Dynamics Simulations of Protein Crystals
Abstract
For decades, X-ray crystallography has been the primary tool used to measure the structure of macromolecules at atomic resolution. As fruitful as this method has been, standard X-ray crystallographic methods often provide a model of only the average structure of macromolecules, ignoring structural heterogeneity and correlated disorder. New methods incorporating heterogeneity and correlated disorder are currently in development, but none have yet been able to overcome the “R-factor gap”: the consistent and significant discrepancy between modeled and predicted crystallographic data. It has been suggested that closing this gap will require more accurate models of ordered solvent and disorder. The work presented here uses molecular dynamics (MD) simulations of protein crystals to investigate models of correlated disorder put forward to predict the “diffuse scattering” (non-Bragg X-ray diffraction) which results from correlated disorder in protein crystals. We also present insights in to the modeling of conformational ensembles, ordered solvent, protonation states, and side-chain disorder gathered from crystalline MD simulations.