/pubchem-finder

Tool for finding similar chemical compounds based on their smiles representation

Primary LanguagePythonMIT LicenseMIT

Suite of tools for indexing and searching similar molecules in Elasticsearch

Tool for finding chemical molecules, compounds, reactions, etc in PubChem (https://pubchem.ncbi.nlm.nih.gov/)

Index pubchem dump into Elasticsearch

docker-compose -f elastic/docker-compose.yml up 
cd pubchem/
export PYTHONPATH=$(pwd)
python3 pubchem.py pull

By default pubchem uses pubchem/ folder in your home directory. You can change this directory using --workdir flag.

--tmpdir should be pointed to SSD. If you already use ssd for --workdir, you can skip this option.

Molecule similarity in Elasticsearch