UC Irvine

The Poisson-Boltzmann Surface Area (PBSA) method, when coupled with molecular dynamics (MM, thus the term MMPBSA) simulations has been widely adopted as an efficient and reliable free energy simulation method to model molecular recognition, such as protein-ligand binding interactions. In this dissertation, we first reviewed the recent developments and applications of the MMPBSA method. This is followed with our analysis of different computational alternatives on their impacts to the agreement of MMPBSA calculations with experiment. Most importantly, we studied the performance of nonpolar solvation models and found that the modern approach that separately models hydrophobic and dispersion interactions dramatically reduces RMSD’s of computed relative binding affinities with respect to experiment. Next, we focus on the quality of the PBSA method in modeling ionic interactions by comparing with both explicit solvent simulations and experiment. Specifically, the molality-dependent chemical potentials for sodium chloride (NaCl) electrolyte were simulated in both SPC/E explicit solvent and PBSA solvent models. Our comparative analysis shows that the molality-dependent chemical potentials of NaCl are reproduced well with both linear PB and nonlinear PB, with nonlinear PB agrees better with SPC/E and experiment. However, for non-electrostatics/van der Waals solvation free energies, the absence of molality dependence in the PBSA method dramatically reduces its quality in modeling of the salt-dependent ionic interactions. Given the above analyses of PBSA method in various applications, we explored a new numerical method for more accurate and robust electrostatic modeling of biomolecular systems. The new method was found to deliver more accurate and better-converged grid potentials than the classical method on or nearby the molecular surface, important for highly demanding applications in direct application to molecular dynamics simulations. Finally, we focus on another important issue to utilize the PBSA method in molecular dynamics simulations. Specifically, we implemented and analyzed a range of standard surface fitting schemes to calculate molecular surface electric fields and dielectric boundary forces, crucial for porting PBSA methods to molecular dynamics engines. Given the developments presented here, we are in a better position to apply PBSA method for biomolecular simulations for more robust free energy simulations with a wider range of applications.