It can be used to cluster the conformations of a molecule in a molecular dynamics trajectory using collection of features. The features could be any quantity as a function of time such as Projections of eigenvector from PCA or dihedral-PCA, distances, angles, channel radius etc.
See details at: gmx_clusterByFeatures homepage
Note
It is developed for GROMACS MD trajectory. However, it can be used with any other trajectory format after converting it to GROMACS format trajectory.
sudo python3 -m pip install gmx-clusterByFeaturesNo dependency on GROMACS. Install and use it.
For more details, visit download and installation section.
| Command | Function |
|---|---|
| cluster | Main module to perform clustering |
| featuresplot | Feature vs Feature plot to check quality of clustering |
| distmat | Distance-matrix related calculations |
| matplot | To visulaize/plot matrix obtained from distmat |
| hole | To calculate cavity/channel radius using HOLE program |
| holeplot | To calculate average and plot hole output radius file |
| holefeatures | To write radius as a features for clustering |
| holeclustersplot | To plot or write radius for clusters separately |
For more details, visit usage section.
Following steps are for development setup. However, these steps can be used for installation using conda
# Clone the repository and setup development environment
git clone --recursive https://github.com/rjdkmr/gmx_clusterByFeatures.git
cd gmx_clusterByFeatures
conda env create -y --prefix ./venv --file environment.yaml # Create development conda environment
conda activate ./venv # Activate the environment
bash -i scripts/build_dev_setup_conda.sh # all build and setup steps are in this script