Library for comparing and benchmarking protein models. In particular it contains an implementation for computing the fraction of native contacts using as a reference a CG pdb.
The dependencies are defined in requirements.txt so it can be install with pip:
pip install git+https://github.com/ClementiGroup/proteka.git
or by cloning the repository and installing localy:
git clone https://github.com/ClementiGroup/proteka.git pip install -e .
Documentation is available here and here are some references on how to work with it.
pip install sphinx sphinx_rtd_theme sphinx-autodoc-typehints
cd docs
sphinx-build -b html source build
This udapte should be done after any update of the main branch so that the documentation is synchronised with the main version of the repository.
git checkout gh-pages
git rebase main
cd docs
sphinx-build -b html source ./
git commit -a
git push