Anthropic Uses Claude to Map Bacteriophage Vault Particle
Anthropic says Claude helped its biology team uncover the structure of a vault like particle made from a bacteriophage protein, as detailed in a company research post. Vaults are large protein shells whose biological function remains unknown.
The researchers used Claude Science to search about 15 billion predicted proteins, design DNA constructs and help test candidate vault proteins. Seven of 15 soluble proteins assembled into vault like particles, while cryo electron microscopy confirmed the shape for four. The team mapped one phage particle built from 96 copies of a protein called L05 and engineered a larger version with about 20 percent more internal volume.
Claude also helped develop an isolation process that reduced preparation time from about three days to 90 minutes. It processed microscopy data using open source software and produced an initial structural model in 34 hours. The project also encountered errors, including an unsuitable growth medium recommendation and a skipped microscopy processing step that produced a flawed map.
The results were released in a preprint. The team has produced the particles only in E. coli and has not confirmed that they form naturally in the bacteria or phages carrying the related genes. Their biological function also remains unknown.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Daily AI Brief.
Also, consider following us on social media:
Daily AI Brief
Daily report covering major AI developments and industry news, with both top stories and complete market updates
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
Claude Designs Smaller Quantum Circuits Across Eight Tasks
Anthropic Automates Search for Training Data That Evades Safety Monitors
Anthropic Expands Free Claude Credits to 20 Science Groups
Anthropic Uses Claude to Verify 123 Open Source Libraries
Claude Security Adds Continuous Repository Scanning
Daily AI Brief: the AI news that matters, in your inbox.