Installation

Requirements

Python 3.7 or newer
RDKit 2020.09 or newer
NetworkX 2.5 or newer

Install with pip

The recommended installation method is via pip. RDKit must already be available in the target environment. It is recommended to use an environment manager (like Conda/Mamba, e.g [Miniforge](https://github.com/conda-forge/miniforge)) and install RDKit via

mamba install -y -c rdkit rdkit

Then install PFASGroups from PyPI:

pip install PFASgroups

Install from source

git clone https://github.com/LucMiaz/PFASGroups.git
cd PFASGroups
pip install -e .

Optional dependencies

Some features require additional packages that are not installed by default:

Feature	Package	Install
PCA / t-SNE	scikit-learn	`pip install scikit-learn`
UMAP	umap-learn	`pip install umap-learn`
Database export	SQLAlchemy	`pip install sqlalchemy`
PostgreSQL export	psycopg2	`pip install psycopg2-binary`

Verify installation

from HalogenGroups import parse_smiles, get_compiled_HalogenGroups

groups = get_compiled_HalogenGroups()
print(f"Loaded {len(groups)} halogen groups")
# Loaded 116 halogen groups

results = parse_smiles(["CCCC(F)(F)F"])
print(results[0].matches[0].group_name)
# perfluoroalkyl

Troubleshooting

ImportError: No module named rdkit

Install RDKit via conda-forge (recommended) or use a wheel from https://www.lfd.uci.edu/~gohlke/pythonlibs/.

ModuleNotFoundError: No module named ‘HalogenGroups’

Make sure you installed the package as PFASgroups (on PyPI the distribution is called PFASgroups; both import HalogenGroups and import PFASGroups work after installation).

scikit-learn / umap-learn not found

These are optional. They are only needed when calling perform_pca(), perform_tsne(), or perform_umap().