Back to home page

Research interests


I am finishing my graduate studies in the Department of Chemistry and Biochemistry at the University of California at San Diego. I work with Dr. J. Andrew McCammon, a talent and inspiration in the field of computational biophysics and biochemistry. The McCammon group uses fundamental theory from physical chemistry, statistical mechanics, quantum mechanics and classical dynamics in conjunction computer models to study biological systems. Specific projects in the group include the characterization of kinase reaction mechanisms, protein-ligand interactions, and synaptic transmission, as well as method development in the all areas pertaining to the simulation of dynamic motions in biomolecular systems.

Free Energy Methodology

My thesis work has focused on the theoretical and methodological development of free energy calculations, which measure the strength of noncovalent molecular associations from molecular dynamics (MD) simulations. The importance of this area prevades computational theory and has been the target of many models. End-point free energy models such as MM/PBSA combine gas-phase, molecular mechanics energies and Poisson Boltzmann/ Surface Area solvation free energies to approximate the free energy of the two .end points. of a binding reaction, the bound and free systems. These models benefit from computational efficiency as only the initial and final states of the system are evaluated and the expensive explicit solvation effects are approximated with implicit solvent approximations. We have presented a theoretical framework for these methods based on statistical mechanics which highlights two issues that had been inconsistently applied in previous analyses; the association free energy, which results from one molecule.s loss of translational and rotational freedom from the standard state, and the conformational free energy due to changes in both molecules. intramolecular motions. We have shown that the association free energy can be accounted for by quantifying the bound ligand.s range of motion during a simulation. Analysis of protein-water, protein-ligand and protein-protein systems suggests that smaller, more loosely bound ligands play a lower entropic penalty than larger, more tightly bound molecules.

Continuum Solvent Models

Agreement between explicit and implicit energies is critical for the success of component based energy models such as MM/PBSA. Energetic compatibility and the accuracy of PB solvation energies in general, are strongly dependent on the continuum parameters that define the solute charge distribution and the boundary between low and high dielectric regions. We are working on optimizing the dielectric boundary definition for the AMBER forcefield based on explicit solvent simulations. We have presented optimized radii for both abrupt and smooth dielectric transitions. The latter are desirable because they decrease grid placement sensitivity and increase energetic convergence. When used in conjunction with an analytic surface definition, smooth dielectric transitions enable continuum force calculations opening the doors to continuum dynamics and minimizations. The accuracy of PB forces is still limited by the quality of the dielectric boundary definition and an area of active research.

Nonpolar Solvation

While the electrostatic contribution to solvation free energies is critical it must be complimented by a non-polar contribution. The most commonly used SASA models assume a linear relationship between the non-polar solvation free energy and the solute.s Solvent Accessible Surface Area (SASA). These crude models are known to be strongly system dependent; they ignore geometric factors such as curvature effects, size dependent phenomena which are expected to change at the 1 nm length scale, and polar-non-polar coupling. We are currently working on a coupled polar-non-polar formalism that collectively accounts for all such phenomena.

Bridging APBS and PMV (making continuum calculations accessible)

In addition to my free energy work, I have also been involved in an NPACI funded alpha project on the integration of the Adaptive Poisson-Boltzmann Solver (APBS) with Python Molecule Viewer (PMV). APBS is a software program that solves the Poisson Boltzmann equation for the electrostatic potential of system in a very computationally efficient manner. PMV, a high power molecular modeling package, will serve as a graphical user interface for the set up and evaluation of APBS calculations. I have been mentoring an undergraduate in the code development, and we recently released the beta version of the APBS-PMV GUI.

Future Aspirations

My strategy for graduate school and future post doctoral work is to develop a solid theoretical foundation in statistical mechanics, biophsyics and biochemistry and a technical foundation in computational modeling and algorithm design. My current roles in method development are in this vein. My larger goal is to apply this foundation to the investigation of medically and/or environmentally relevant biological systems, exploring areas such as aging and disease for the former and bioremediation, pollution control, and renewable energy production for the latter.