It is a ongoing challenging to determine the low variance, low bias simulation set-ups to perform accurate and reliable free energy calculations for solution phase thermodynamic properties. We use multistate reweighting techniques to rapidly evaluate thousands of possible combinations of simulation parameters, such as PME parameters, cutoffs, and switch distances for series of benchmark systems.  This process allows easy identification of exactly what speed tradeoffs must be made to reach a given level of precision in parameter space.  Interestingly, we find that we can estimate differences of free energy differences much more efficiently than we can compute the free energies differences themselves when dealing with states far away in phase space. We also present recent work on validation of error estimates for robust and reliable free energy simulations.

Validation of free energy calculation methods: Thermodynamic sampling of large parameter spaces using multistate reweighting methods

244th National Meeting (2012)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.