AdaptiveMasterMSMΒΆ
AdaptiveMasterMSM is an open-source implementation of adaptive sampling algorithms exploiting markov state models based on the rate matrix.
The AdaptiveMasterMSM python package is designed to carry out adaptive sampling unbiased simulations in order to cover slow biological time scales. Our implementation uses Gromacs and multiprocessing to simulate parallel short trajectories. Markov state models are employed between consecutive trajectories to determine where best to sample from. The latter is decided according to (1) low-counts and (2) populations metrics, despite more options are being investigated. In contrast to other similar computer programs, AdaptiveMasterMSM employs the MasterMSM python package to construct the Markov state models, which allows to work with the rate matrix instead of the usually employed transition matrix.
