LIT-PCBA
Literature-curated dose-response PubChem BioAssay set; 15 targets with unbiased actives/inactives.
Composite
80.8
Experimental validation
Retrospective
Stages
Hit ID
Modalities
small-molecule
Task types
virtual-screening
Size
targets: 15
actives: 7,844
inactives: 2,533,936
actives: 7,844
inactives: 2,533,936
License
CC-BY
First release
2020-03
Last updated
2023-06
Official site
Leaderboard
→ leaderboard
Dataset
Code / GitHub
HuggingFace
→ HF
Paper
LIT-PCBA: An Unbiased Data Set for Machine Learning and Virtual Screening · Tran-Nguyen VK, Jacquemard C, Rognan D · 2020 · paper · doi:10.1021/acs.jcim.0c00155 · 310 citations
Flags
none
Experts
Groups
Hosted by
—
Related benchmarks
Rubric (7-criterion)
rigor
5
coverage
3
maintenance
2
adoption
4
quality
5
accessibility
5
industry_relevance
4
Notes
Much fairer than DUD-E; small target count limits coverage.