UC Davis-led team designs Rubisco de novo with protein language models (bioRxiv)
KevinKaichuang · x · 2026-10-09
A bioRxiv preprint, "De novo Rubisco design with protein language models," from UC Davis, Lawrence Berkeley National Laboratory, Joint BioEnergy Institute and collaborators (including NVIDIA's Simon Chu), applies protein language models to de novo design of Rubisco, the key photosynthetic enzyme. First author Alex Kehl will present the work at an ML4PE seminar on October 13, 2026.
More from Research
- COLM2026 talk: LLM factual generation-verification gaps evolve across fact lifecycle — caglarml · 2026-10-09
- Adaption AI tackles unverifiable-domain data evals with an agentic checklist — sarahookr · 2026-10-09
- Mathematician dismisses AI proof mismatch flap: NL-vs-Lean gap is just a missed verification step — AlexKontorovich · 2026-10-09
- Cryptographer Matthew Green: losing public-key crypto means weaker standards, not doom — matthew_d_green · 2026-10-09
- AI-assisted massive review of negative weights in difference-in-difference launched — RexDouglass · 2026-10-09
- LightOnOCR-3 training: using OCR reference text to guide logical block grouping — IgorCarron · 2026-10-09