Published June 17, 2025
| Version v2
Dataset
Open
Datasets, alphabets and models from paper 'Reverse Engineering Molecules from Fingerprints through Deterministic Enumeration and Generative Models.
Authors/Creators
-
1.
National Research Institute For Agriculture, Food And Environment
-
2.
University of Manchester
Description
Files utilized and produced within the molecule-signature project:
alphabets.zip: Alphabets of molecule signatures built from MetaNetX, eMolecules and ChEMBLdatasets.zip: Datasets from MetaNetX, eMolecules, DrugBank and MolForge used to build alphabets, train generative models, and evaluate methods.models.zip: PyTorch/Lightning models and SentencePiece tokenization models for decoding SMILES from ECFP.
See embedded README.md files and the publication for in depth details.
Files
alphabets.zip
Additional details
Related works
- Is supplement to
- Software: https://github.com/brsynth/molecule-signature-paper (URL)
- Software: https://github.com/brsynth/molecule-signature (URL)
Funding
- Agence Nationale de la Recherche
- Galaxy-BioProd - Galaxy-BioProd: An operating portal for the production of biosourced products ANR-22-PEBB-0008
- Agence Nationale de la Recherche
- GENCI - GENCI ANR-17-EQPX-0001
- Agence Nationale de la Recherche
- IFB (ex Renabi-IFB) - Institut français de bioinformatique ANR-11-INBS-0013
Software
- Repository URL
- https://github.com/brsynth/molecule-signature-paper
- Programming language
- Python
- Development Status
- Active