Comparison of free energy methods for molecular systems

We present a detailed comparison of computational efficiency and precisionfor several free energy difference ($\Delta F$) methods. The analysis includesboth equilibrium and non-equilibrium approaches, and distinguishes betweenuni-directional and bi-directional methodologies. We are primarily interestedin comparing two recently proposed approaches, adaptive integration andsingle-ensemble path sampling, to more established methodologies. As testcases, we study relative solvation free energies, of large changes to the sizeor charge of a Lennard-Jones particle in explicit water. The results show that,for the systems used in this study, both adaptive integration and path samplingoffer unique advantages over the more traditional approaches. Specifically,adaptive integration is found to provide very precise long-simulation $\DeltaF$ estimates as compared to other methods used in this report, while alsooffering rapid estimation of $\Delta F$. The results demonstrate that theadaptive integration approach is the best overall method for the systemsstudied here. The single-ensemble path sampling approach is found to besuperior to ordinary Jarzynski averaging for the uni-directional,``fast-growth'' non-equilibrium case. Closer examination of the path samplingapproach on a two-dimensional system suggests it may be the overall method ofchoice when conformational sampling barriers are high. However, it appears thatthe free energy landscapes for the systems used in this study have rathermodest configurational sampling barriers.Comment: 13 pages, 3 figure