SSBP and GSBP
in which all atomic and structural details of the solvent
molecules are ignored (as in PB calculations) may not always
be desirable. In some cases, it may be advantageous to use
an intermediate approach which consists in keeping a small
number of explicit solvent molecules in the vicinity of the
solute, and representing the remaining bulk with an effective
solvent boundary potential.
Separating the multidimensional solute-solvent configurational integral in terms of "inner'' solvent molecules nearest to an arbitrary solute, and the remaining "outer'' bulk solvent molecules, we showed that the solvent boundary potential corresponds to the solvation free energy of an effective cluster comprising the solute and inner explicit solvent molecules embedded in a large hard sphere (Beglov and Roux, 1994). The hard sphere corresponds to a configurational restriction on the outer bulk solvent molecules; its radius is variable, such that it includes the most distant inner solvent molecule. An approximate Spherical Solvent Boundary Potential, called SSBP, based on this formulation was shown to yield accurate results in computer simulations. SSBP is meant to simulate solutes surrounded by bulk isotropic solvent. To simulate a small region of a protein or active site, the Generalized Solvent Boundary Potential, called GSBP, was developed (Im et al, 2001). To properly equilibrate the finite GSBP region, a grand canonical monte carlo method, called GCMC, was developed.
GSBP with GCMC is a powerfull strategy for computing absolute binding free energy of ligands in buried sites (Deng and Roux, 2008).
All these methods have been implemented in CHARMM.