Comparison between integrated and parallel tempering methods in enhanced sampling simulations.

Recently, we introduced an integrated tempering approach to enhance sampling in the energy and configuration space for large systems. In this paper, we show that this new method has a higher efficiency than bias potential and generalized ensemble methods, such as accelerated molecular dynamics and replica-exchange molecular dynamics (parallel tempering) methods, in yielding thermodynamic averages. Particularly, the sampling efficiencies in both energy and configuration spaces are compared in details between integrated and parallel tempering methods. Related issues regarding the efficiency involved in the usage of the parallel tempering method are also discussed.