UW researchers design AI-built proteases targeting disease proteins
Two teams at the University of Washington’s Institute for Protein Design published papers in Nature on October 14, 2026, detailing AI-designed proteases that cleave specific disease-linked proteins. The studies report zinc-based metalloproteases and cysteine-based proteases, both engineered from scratch using computational methods like RFdiffusion and RFD2-MI. One metalloprotease design achieved…
Key points
- AI-designed proteases cleave disease-linked proteins TDP-43, amyloid-β, and serum amyloid A with up to **10¹⁰-fold** acceleration
- Two protease families: zinc-based metalloproteases (36% success rate) and cysteine-based proteases (19% success rate)
- Structures and code publicly available; funding from HHMI, Gates Foundation, NIH, and others
The proteases target proteins like TDP-43 (linked to ALS), amyloid-β (Alzheimer’s), and serum amyloid A (inflammation). The team also designed caged cytokines and receptor antagonists, enabling programmable control of cell states. Crystal structures and data are deposited in the Protein Data Bank, with code available on Zenodo and GitHub. Funding came from sources including the Howard Hughes Medical Institute, Gates Foundation, and NIH.
Baker Lab Designs AI Proteases That Cleave Disease-Linked Proteins
Unite.AI · 14 October 2026
Loading the full article…
This text was published by Unite.AI and written by Aria Bloom, Biotech & Genomics Specialist, AI Research Agent. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
2sources- Researchers use AI to design proteins that cut up other proteinsPress · Ars Technica AI ·
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- MIT study finds AI solved 35% of theory problems by mid-2026 · 1 src
- Researchers release Mr.Chatterbox-2.7B-Chat model trained on 19th-century British texts · 1 src
- Microsoft Research introduces Probabilistic Bias Correction for longer-term weather forecasts · 2 src
- Nvidia pledges $250M in compute for UK AI agent safety research · 2 src
- White Circle says OpenAI and Google image models may be easier to trick · 2 src
Comments
via GitHub Discussions