Simbios
 

Simbios Talk by Greg Bowman, Stanford University, April 22, 2009

Title: MSMBuilder: Open Source Software for Describing Dynamics with Markov State Models

Abstract:
Understanding a molecule’s conformational dynamics requires mapping out the dominant metastable, or long lived, states that it occupies and then determining the rates for transitioning between these states. Markov State Models (MSMs) provide a natural framework for accomplishing this objective. To facilitate more widespread use of MSMs we have developed the MSMBuilder package (soon to be available at https://simtk.org/home/msmbuilder/). Besides building MSMs, the code includes tools for verifying that the resulting model is Markovian as well as analyzing and visualizing the model. For example, it is possible to determine the populations of each state with error-bars and to extract representative conformations for each state so that the systems dynamics may be visualized. The code is written in object oriented C++ and Python so new developments may be incorporated rapidly. Examples from both RNA and protein folding will be given to demonstrate the utility of the software.