It's not completely clear what you're trying to do, and why you want to separate the components. Binding energy can be difficult to compute; getting the potential energy of an interaction is fairly easy in CHARMM, and does not require separating the components. The difficult part is estimating the entropy contribution, the TdS term in the standard free energy expression. Part of the problem is sampling; most simulations are too short to explore a major fraction of the phase space of a binding interaction for two large, flexible molecules.

Rick Venable
computational chemist