Claude Designs Protein Binders Against Scorpion Toxins

Claude spent 64 hours designing proteins against scorpion venom.

New on Anthropic’s Science Blog: designing protein binders against scorpion neurotoxins. Scorpions sting about 1.2m people every year, Anthropic says, and antivenoms are often costly or in short supply. “Claude has made an early step toward finding new treatments.”

Working with @citrini, Anthropic asked Claude to design proteins that bind to five scorpion toxins, covering the part of the toxin that latches onto cells.

The 17 binders bound to dangerous toxins produced by scorpions from Mexico, North Africa and the Middle East. They did not bind to the toxin of a species from Brazil.

Exploratory work, with the data released.

Anthropic is careful about the claim. It calls this exploratory work and says it is not claiming to have produced an antivenom. That would require testing the binders within living cells, then in mice, among many other steps. It calls the project an example of how Claude can concretely assist research in protein design.

Anthropic is releasing all 211 of the tested designs and the lab data as a resource for scientists, along with a full technical report.

“The beginning of autonomous AI scientists.”

@kimmonismus posted the story two hours before Anthropic’s thread. He describes the work as done by Anthropic’s internal Claude research model, which “carried out much of a research project on its own, with results confirmed in the lab.”

Given a starting prompt and four follow-ups, he writes, Claude used specialized tools to generate more than 318,000 protein designs in 64 hours. It chose where they should attach to scorpion toxins and selected candidates for testing. He notes the 211 tested candidates included additions made with human help.

In his account, Claude handled a substantial stretch of research: choosing an approach, coordinating tools and deciding what deserved a physical experiment. “From here on it only gets more autonomous,” he writes.