Arc Institute uses genome language models and AI agents to mine the genome's dark matter for functional RNAs
Afinetheorem · x · 2026-10-23
Arc Institute released a landmark preprint using genome language models and AI agents for biological discovery and genome mining. The non-coding DNA between genes—often called the genome's 'dark matter'—encodes RNAs that sense metabolites, regulate genes, and underpin tools like CRISPR. These RNAs are hard to find with conventional tools because structure and function persist even as sequences diverge. Patrick Collison highlights it as deep learning applied to a new kind of knowledge, and a preview of agents accelerating discovery in vast combinatorial spaces with wet-lab biology.
Related event: Arc Institute Releases COMPASS to Mine Genomic "Dark Matter" with AI(4 posts)→
More from Research
- Benchmarking AI agents in biology takes center stage at Genentech — kenbwork · 2026-10-23
- Torus wave visualizations show how topology discretizes continuous vibrations into exact eigenmodes — forestmars · 2026-10-23
- Peking University and Alibaba unveil OmniDR-Bench for video-centered deep research agents — Qwen · 2026-10-23
- NVIDIA's Foundation4D unifies AV perception and reconstruction, topping Waymo and nuScenes — nvidia · 2026-10-23
- NAVER AI Lab releases LaLa-World, scaling latent world models with natural-language actions — naver-ai · 2026-10-23
- FAPL From HKUST and Tencent Makes Video DiTs Actually Use MLLM Latents — tencent · 2026-10-23