MoleculeACE
Benchmark testing model robustness on activity cliffs across 30 ChEMBL targets.
Composite score: 83.3
Rubric (1–5 per criterion)
rigor
5/5
coverage
3/5
maintenance
3/5
adoption
4/5
quality
5/5
accessibility
5/5
industry_relevance
4/5
Metadata
Stages
Lead ID / ADMET
Modalities
small-molecule
Task types
regressionactivity-cliff
License
MIT
First release
2022-11
Last updated
2024-05
Flags
none
Size & scope
- targets: 30
- molecules: 48000
Primary paper
Title
Exposing the Limitations of Molecular Machine Learning with Activity Cliffs
Authors
van Tilborg D, Alenicheva A, Grisoni F
Year
2022
DOI / arXiv
Citations
180
Links
- Official site: https://github.com/molML/MoleculeACE
- GitHub: https://github.com/molML/MoleculeACE
- Leaderboard: N/A
Hosted by (initiatives)
- not hosted by any tracked initiative
Experts (primary authors / maintainers)
Groups (host labs / companies / consortia)
Related benchmarks
Notes (honest caveats)
Critical stress-test for generalization; exposed GNN weaknesses.