Particle.news
Download on the App Store
5 ARTICLES
·
4w ago

AI-Driven Virtual Lab Designs and Validates SARS-CoV-2 Nanobodies

Operating with less than 1% human intervention, the platform is being adapted to tackle diverse biomedical and data-analysis challenges.

Scientists Wesley Wu, Nash Bulaong, and John Pak (from left to right), pictured at their lab in San Francisco on July 25, 2025, all contributed to the Virtual Lab paper showing how a group of AI scientists can contribute to original research. (CZ Biohub San Francisco)
A message reading "AI artificial intelligence", a keyboard, and robot hands are seen in this illustration taken January 27, 2025. REUTERS/Dado Ruvic/Illustration
Image
Image

Overview

  • The study published July 29, 2025 in Nature by Stanford and Chan Zuckerberg Biohub researchers outlines an autonomous Virtual Lab composed of AI agents with distinct scientific roles.
  • An AI Principal Investigator agent autonomously assembled immunology, computational biology, and machine learning specialists to generate dozens of nanobody candidates against SARS-CoV-2 within days.
  • Laboratory tests confirmed the AI-designed nanobodies were structurally stable, bound tightly to spike proteins of recent variants, and exhibited no off-target binding.
  • Human oversight was limited to about 1% of the workflow, with all AI meetings and decisions logged via transcripts for full transparency.
  • Researchers are now extending the Virtual Lab framework to broader biomedical problems and deploying specialized data-analysis agents to reevaluate existing research findings.