Harvard lab uses Biomni AI agents to decode the hidden "grammar" of DNA
anshulkundaje · x · 2026-11-22
A science blog post from the Phylo Academic Lab Program details how the Naqvi Lab at Boston Children's Hospital and Harvard Medical School used Biomni Lab AI agents to study why neural crest cells develop into very different tissues — facial bones vs. peripheral nerves — despite sharing the same DNA and key regulatory proteins.
- The agents processed genomic data and ran deep learning models to generate thousands of predictions on how DNA binding sites regulate gene activity alone and in pairs
- Finding: most DNA "grammar" rules are shared across neural crest cell populations
A concrete case of AI agents accelerating the scientific discovery loop.
More from AGI Musings
- DHH: AI is bringing back the individual hacker's power to shape computing — a16z · 2026-11-22
- Mathematicians begin digesting the now-verified proof: standard pieces, cleverly composed — PMinervini · 2026-11-22
- Erik Hoel defends 'qualiaphobia': the moral case for engineering non-conscious AI — erikphoel · 2026-11-22
- Colgate says LLM role-play can generate ~90% accurate survey data, challenging market research — ZeroStateReflex · 2026-11-22
- teortaxesTex cites Yudkowsky's Riemann passage to contest the 'RH-capable AI can kill you' threat model — teortaxesTex · 2026-11-22
- AlphaChip and three TPU generations: a mine that digs itself is a throne — krishnan · 2026-11-22