Utility to run the NPC APIs to classify SMILES, plus preprocessed datasets.
Using this utility, we have already labelled SMILES from the following datasets which we share on Zenodo.
The datasets are stored in a gzip-ed JSON file with the following format:
[
{
"class_results": [
"Cyclic peptides",
"Microcystins"
],
"superclass_results": [
"Oligopeptides"
],
"pathway_results": [
"Amino acids and Peptides"
],
"isglycoside": false,
"smiles": "CC(C=CC1NC(=O)C(CCCN=C(N)N)NC(=O)C(C)C(C(=O)O)NC(=O)C(CC(C)C)=NC(=O)C(C)NC(=O)C(C)N(C)C(=O)CCC(C(=O)O)NC(=O)C1C)=CC(C)C(O)Cc1ccccc1"
},
{
"class_results": [
"Cyclic peptides",
"Depsipeptides"
],
"superclass_results": [
"Oligopeptides"
],
"pathway_results": [
"Amino acids and Peptides",
"Polyketides"
],
"isglycoside": false,
"smiles": "CC(=O)OC1c2nc(cs2)C(=O)OC(CCCC(C)(Cl)[37Cl])C(C)C(=O)OC(C(C)(C)O)c2nc(cs2)C(=O)OC1(C)C"
}
]First, clone this repository:
git clone https://github.com/LucaCappelletti94/npc-labeler.gitNavigate in it and install the requirements:
cd npc-labeler
pip install -r requirements.txtThen, you can run the labeler by providing the input file and the output file:
python3 labeler.py --input <input_file> --output <output_file>For instance, suppose you want to classify the SMILES in the metadata of an MGF document and store it into a classified_matchms.json.gz file. You can do it by running:
python3 labeler.py --input matchms.mgf --output classified_matchms.json.gzSimilarly, for a SSV file:
python3 labeler.py --input CID-SMILES.ssv --output pubchem.json.gz