Monte Carlo method to obtain diffusion constants
mcdiff is a Python program for obtaining diffusion and free energy profiles from molecular dynamics trajectories. It uses Bayesian Analysis and Monte-Carlo importance sampling to determine those profiles and in turn calculate membrane permeabilities via the Inhomogeneous Solubility Diffusion model.
mcdiff uses the following third-party libraries
- numpy >= 1.16.0
- scipy >= 1.2.0
- matplotlib >= 2.0.2
Installing the code takes approximately 5 minutes.
- Linux or MacOS
- python >= 3.5
Installation via anaconda
conda install "numpy>=1.16.0" "scipy>=1.2.0" "matplotlib>=2.0.2"
or pip
pip install "numpy>=1.16.0" "scipy>=1.2.0" "matplotlib>=2.0.2"
git clone https://github.com/annekegh/mcdiff.git
cd mcdiff
python setup.py install
The setup process installs the following terminal commands
run-mcdiff
The main command for running the Bayesian Analysis. For seeing the available options,
type run-mcdiff --help. For examples on how to create and format the transition matrix,
take a look into the examples directory.
mcdiff-infty
A command to extrapolate profiles to infinite lag time.
plotresults
A command to plot the final profiles.
MIT License. See LICENSE file.
- An Ghysels, Ghent University
- Gerhard Hummer, MPI for Biophysics