HARMONSMILE: Harmonize SMILES Strings for Cheminformatics and Machine Learning
Description
v0.3.2 (2026-07-09)
HARMONSMILE is a cheminformatics toolkit for preparing and harmonizing molecular structure tables across PubChem, ChEMBL, and independent SMILES datasets. It provides deterministic table loading, robust identifier handling, RDKit-based SMILES canonicalization, and an auditable harmonization layer for downstream curation, comparison, and machine learning workflows. HARMONSMILE preserves source SMILES, emits a canonical/isomeric/Kekulé RDKit representation for compatibility and deduplication, and produces a harmonized canonical/isomeric/aromatic SMILES representation using controlled RDKit-native standardization policies.
Notes
Files
NanoBiostructuresRG/harmonsmile-v0.3.2.zip
Files
(472.6 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:fbd916f3d366e0b19ddd41b990b7aa53
|
472.6 kB | Preview Download |
Additional details
Related works
- Is supplement to
- Software: https://github.com/NanoBiostructuresRG/harmonsmile/tree/v0.3.2 (URL)
Dates
- Available
-
2025-07-09
Software
- Repository URL
- https://pypi.org/project/harmonsmile/
- Programming language
- Python
- Development Status
- Active