COMPASS Model Improves Genetic Perturbation Response Prediction
anshulkundaje · x · 2026-08-28
Rohit Singh highlights research on perturbation prediction via the COMPASS model. The study addresses the paradox where deep learning models often lose to simple baseline averages in predicting perturbation responses. Analyzing 2,270 CRISPRi perturbations across six cell lines, it reveals a conserved organization in responses and proposes an interpretable linear model to decompose shared and gene-specific components.
More from Research
- The Hundred-Page Language Models Book Released — burkov · 2026-08-28
- Bittensor humanoid robot competition yields superhuman long jump skills — markjeffrey · 2026-08-28
- Research shows injecting 1,000 fingerprints into open models causes almost no performance drop — 0xsachi · 2026-08-28
- AI safety researchers concerned multi-agent training may make models more helpful to AIs than humans — dhadfieldmenell · 2026-08-28
- Qwen4 speculation: 27B model may see a leap in intelligence — Iory1998 · 2026-08-28
- LLMs can't originate a KRAS-style insight, but excel at cross-domain hypothesis synthesis — Ghost_Pilot_MD · 2026-08-28