Cell fate predictions need not reconstruct every gene, given noisy single-cell data
anshulkundaje · x · 2026-10-10
Anshul Kundaje agrees with Fabian Theis that cell fate prediction is a much better-defined objective than raw perturbation effect prediction: given how poor single-cell measurements are, one could in principle get correct answers without perfectly predicting every gene's expression. This post substantially duplicates other fragments of the same thread.
Related event: Theis and Kundaje Debate Phenotype-First Perturbation Modeling(4 posts)→
More from Research
- ThunderSyncRL speeds up synchronous agentic RL by up to 1.9x with zero policy staleness — StanfordAILab · 2026-10-10
- Prompt injection detection benchmarks have badly mislabeled data, researcher finds — evilsocket · 2026-10-10
- Researcher Replicates Claude 4's 'Bliss Attractor' Eval: LLMs Fall Into Distinct Obsession Basins — rgblong · 2026-10-10
- YC-backed team builds 8,000 sqft wet-lab in 30 days, AI compounds kill $13B pest — ycombinator · 2026-10-10
- De novo glycoprotein design: glycans become a designable parameter in new preprint — DaveJuergens · 2026-10-10
- LeWAM: end-to-end JEPA world model plans 32x faster, lifts robot success to 89.7% — randall_balestr · 2026-10-10