Protein language model redesigns entire helix — but why it works remains a mystery
sim0nsays · x · 2026-09-12
An AI-for-science researcher shares that a protein language model (pLM) redesigned most of a helix and introduced other mutations — and the design works. But understanding why it works would require experimental effort comparable to the entire project, echoing Feynman's famous quote. The author adds that the figure was meant to be about AI in biology, but now speaks to every field: models deliver working solutions whose mechanics we can't yet explain.
More from Research
- Continued pretraining vs RAG: an accuracy and performance comparison on Qwen 3.5 4B — funJS · 2026-09-12
- Researcher speculates spatial reasoning leap comes from Blender training data — yoavartzi · 2026-09-12
- tszzl pushes back on Fermi paper: 50-OOM lognormal abiogenesis prior under-justified — tszzl · 2026-09-12
- Continued pretraining vs RAG on Qwen 3.5 4B: a hands-on accuracy and performance comparison — funJS · 2026-09-12
- OpenAI launches GPT-Rosalind for biological reasoning in API and Codex — OpenAIDevs · 2026-09-12
- MIT quantum lab uses GPT-5.6 Sol and Codex to automate chip measurements — OpenAI · 2026-09-12