Skip to content

jdrusso/msm_we

Repository files navigation

msm_we

https://github.com/jdrusso/msm_we/actions/workflows/mamba-test.yml/badge.svg?branch=main&event=push Documentation Status https://codecov.io/github/jdrusso/msm_we/branch/main/graph/badge.svg?token=BGVJ3BY6S2

Background

This is a package for doing history-augmented MSM (haMSM) analysis on weighted ensemble trajectories.

Weighted ensemble data produced from simulations with recycling boundary conditions are naturally in a directional ensemble. This means that a history label can be assigned to every trajectory, and an haMSM can be constructed.

Features

  • Compute a history-augmented Markov state model from WESTPA weighted ensemble data
  • Estimate steady-state distributions
  • Estimate flux profiles
  • Estimate committors
  • WESTPA plugins to automate haMSM construction
  • WESTPA plugin to automate bin+allocation optimization

Known Issues

  • Due to H5py version dependencies, this is currently not compatible with Python 3.10.
  • Sometimes, on Python3.7 (and maybe below) the subprocess calls will fail. This may manifest as a silent failure, followed by hanging (which is very fun to debug!) To fix this, upgrade to Python 3.8+.
  • If running with $OMP_NUM_THREADS > 1, Ray parallelism may occasionally silently hang during clustering / fluxmatrix calculations

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

About

History-augmented Markov analysis of weighted ensemble trajectories.

Topics

Resources

License

Stars

Watchers

Forks

Contributors 4

  •  
  •  
  •  
  •  

Languages