A Novel Algorithm For Creating Density Dependent Implicit Solvent Models

Implicit solvent simulations, i.e. those in which solvent molecules are not explicitly simulated and the solute-solute potential is modified to incorporate this solvent effect, have the potential to speed up many relevant molecular simulations, including studies of proteins and surfactants. While a pairwise potential between solute sites is simplest, the additional parameterization based on the local solute concentration has the potential to increase the accuracy of the simulations with only a marginal increase in computational cost. We describe here a method in which the radial distribution function and free energy of solvation of a system of interest are first measured in a fully explicit solvent simulation, and then reproduced across a range of densities in an implicit solvent simulation.