BIBEE: a Rigorous and Computationally Efficient Approximation to Continuum Electrostatics

Abstract

The computational costs associated with modeling biomolecular electrostatics using continuum theory have motivated numerous approximations, such as Generalized-Born (GB) models, that can be computed in much less time. Unfortunately, most of these approximate models abandon physics in favor of computational efficiency. On the other hand, a new approximation method for molecular electrostatics, called BIBEE (boundary-integral-based electrostatics estimation), retains the underlying physics of continuum theory, but is nearly as efficient as Generalized-Born models. The BIBEE approach derives from well-known results in potential theory and the theory of boundary-integral equations. Three main results demonstrate the value BIBEE may hold for biomolecular analysis and design. First, the integral-equation theory clarifies the origin of accuracy of the Coulomb-field approximation (CFA). Second, BIBEE models offer significantly better accuracy for individual pairwise interactions, relative to GB methods. Third, BIBEE readily provides provable upper and lower bounds to the electrostatic solvation free energy of the original (exact) continuum-theory problem.