bioRxivJClub
JUMP Cell Painting chemical dataset: Morphological map of 115,796 compounds in human cells
bioRxiv · · Preprint · Open access
Chandrasekaran, Seal + 41 more
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
JUMP Cell Painting generated 1.53 billion single-cell profiles for 115,796 compounds in human osteosarcoma cells, enabling a validated public morphological reference despite lab variability.
- Why it matters: A major gap has been the lack of a large, public chemical perturbation dataset to benchmark and support image-based profiling for biological and therapeutic discovery.
- What they did: The JUMP-Cell Painting Consortium produced Cell Painting images across 12 laboratories from 18 partners, then quantified phenotypic activity, reproducibility, and profile similarity for compound perturbations.
- The result: Over 15% of compounds show statistically significant phenotypic activity, reproducibility declines by 34% versus single-laboratory data, and target-sharing compounds often have similar image-based profiles.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.
- Open access
- 98 cites