cramg
commented
3 years ago A review of the MM/PB(GB)SA methods along with MD simulations is presented. Hou et al. mentioned that these methods are sensitive to simulation protocols. They discussed, mainly, about the effect of the length of MD simulations (in some cases, a short simulation can assess better results) , entropy analysis (for a single trajectory approach, it is more convenient to not consider this term), solute dielectric constant (they proposed a measurement called PSASAD to guide the chose of this value) and comparison of the efficienty of different PB and GB models (for relative binding energies, GB model performs better, but in absolute energy calculations, PB approach is more rigorous).
Hou, Tingjun, Junmei Wang, Youyong Li, and Wei Wang. “Assessing the Performance of the MM/PBSA and MM/GBSA Methods. 1. The Accuracy of Binding Free Energy Calculations Based on Molecular Dynamics Simulations.” Journal of Chemical Information and Modeling 51, no. 1 (2011): 69–82.