Extracting Free Energy Landscapes using Markov State Models

Markov state models (MSMs) have found widespread use in interpreting conformation changes of proteins and understanding the complex molecular rearrangements necessary for drug delivery. By discretizing structural data from molecular simulations into several stable and metastable states, fundamental processes underlying a larger process may be elucidated. Here, we explore the use of Markov State Models in constructing free energy landscapes defined on important collective variables. In particular, we explore particle transport applications through porous media where the MSM is utilized within an unbiased advanced sampling method to obtain effective free energies and elucidate the fundamental processes limiting transport. The resulting method is efficient, accurate, and enables massively parallel data acquisition and simple error estimation. We will further discuss extensions of the method to driven, nonequilibrium processes such as solute nanofiltration.