Anthropic: Claude Autonomously Designs Protein Binders, 14/15 Targets Succeeded
rohanpaul_ai · x · 2026-08-19
Anthropic published one of its most ambitious scientific experiments with Claude: given only a detailed expert-written protocol, Claude autonomously ran entire computational protein-design campaigns — researching targets, choosing binding sites, installing and running open-source protein-design tools, generating and filtering candidates, and selecting final proteins for lab testing. Humans made no individual design decisions. Working with Adaptyv Bio and Twist Bioscience, the designs were independently built and tested: out of 1,320 designs with usable measurements, 354 worked in the lab. This could shift the bottleneck from manually designing thousands of candidates to experimentally testing a much smaller AI-selected set.
Related event: Anthropic Says Claude Designed Protein Binders for 14 of 15 Targets(45 posts)→
More from Research
- UCSD Researchers Propose Dataset Watermarking Detectable via Black-Box API Access — yuxiangw_cs · 2026-10-16
- EmbodiedEvo tops RoboDojo as agents inherit reusable skills and lessons — Aiden_Tech_Ai · 2026-10-16
- CHI reviewer laments: papers' Discussion sections now cite nothing — IanArawjo · 2026-10-16
- Casper: Understanding 90% of neurons ≠ understanding the network — StephenLCasper · 2026-10-16
- NVIDIA Launches cuPotential in CUDA-X to Bring AI Agents to Semiconductor Materials Simulation — tomaszbednarz · 2026-10-16
- Agentic Search Systems Are Rewarded Only for Final Answers, Raising Faithfulness Concerns — burkov · 2026-10-16