New Approaches to Accurate Predictions of Free Energies: From Proteins to Organic Nanostructures

Citation

Pascal, Tod Augustin (2010) New Approaches to Accurate Predictions of Free Energies: From Proteins to Organic Nanostructures. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/P0JH-QM60. https://resolver.caltech.edu/CaltechTHESIS:10262009-204842274

Abstract

Although computer simulations have been applied to the calculation of solvation free energies for a wide range of molecules, reliable calculations including explicit consideration of entropy and quantum effects (zero-point energy corrections) are less common, owing to the enormous computational effort required by standard perturbation methods. Accurate calculations of entropies are essential if computer simulations are to become more useful tools for obtaining molecular insights into solvation and ligand binding phenomena.

We report on the extension of a method of calculating exact entropies and quantum effects from standard MD simulations. This novel method is applied to the investigation of three model cases: 1) the folding of a DNA three-way junction 2) the efficacy of binding in a protein-protein interaction, critical in the pathogenesis of bacterial meningitis in neonates and 3) the free energy of water molecules at two extreme surfaces, as model systems for an investigation of the hydrophobic effect. We develop a scheme to partition the free energy into the per-atom components, and show that the water molecules in the first hydration shell have unique character. Understanding their nature is critical to discovering the underlying physics in these systems.

Thesis Files