Arc Institute Builds High-Quality Perturb-seq Data for the 2026 Virtual Cell Challenge
anshulkundaje · x · 2026-10-02
Arc Institute details its effort to generate high-quality Perturb-seq data for the 2026 Virtual Cell Challenge, noting that in biology compute has scaled but data has not.
Key points:
- Virtual cell models aim to predict effects of genetic mutations, environmental changes, and small molecules—but need large-scale causal, single-cell resolution data to make trustworthy predictions;
- The 2025 challenge was few-shot (predict unseen perturbations in one cell line); 2026 leaps to zero-shot predictions across six cell lines spanning multiple axes of biological diversity;
- Contestants receive transcriptional profiles of unperturbed cells as input, and scoring depends on accurate ground-truth perturbation data per cellular context;
- Arc positions the datasets as a growing community resource for the field.
More from Research
- Researcher: arXiv flooded with slop, ML academia increasingly marginalized — joshua_saxe · 2026-10-02
- Meta paper: a dedicated controller lifts long agent runs from 63.7% to 71.5% at same budget — dair_ai · 2026-10-02
- Robert Long's open questions on introspection in empirical AI welfare — rgblong · 2026-10-02
- DeepMind researcher: billions of recursive agents will soon run over mathematicians — CsabaSzepesvari · 2026-10-02
- Surgical Data Science Collective convenes experts on whether AI truly understands surgery — ddonoho · 2026-10-02
- NVIDIA and Waterloo release PixelUMM: an encoder-free multimodal model reading and writing raw pixels — CSProfKGD · 2026-10-02