Stanford researchers use generative AI to design functional synthetic viruses

generative AI synthetic biology bacteriophages E. coli biosafety

Researchers from Stanford University and the Arc Institute have achieved a milestone in synthetic biology by using generative AI to design complete, functional viral genomes from scratch. Using genomic language models named Evo 1 and Evo 2, the team generated approximately 700,000 potential viral designs. Following computational filtering and chemical synthesis, 16 synthetic bacteriophages were successfully tested in a laboratory setting.

These artificial viruses were specifically designed to target Escherichia coli (E. coli) bacteria. In laboratory tests, the synthetic viruses demonstrated the ability to infect, replicate, and eliminate bacterial colonies, including strains that had developed resistance to natural viruses. The research, published in the journal Science, suggests potential medical applications, such as phage therapy to combat antibiotic-resistant superbugs.

However, the breakthrough has also triggered significant biosafety and biosecurity concerns. Experts have warned that the ability to generate viable biological agents via AI could be misused to create pathogens. To mitigate immediate risks, the researchers noted that human-infecting, animal-infecting, plant-infecting, and fungi-infecting sequences were excluded from the model's training data.

Entities: Arc Institute · Brian Hie · Brian L. Hie · EVO · Escherichia coli · Gates Foundation · Harvard University · Samuel H. King · Samuel King · Stanford University · Stephen Elledge · viral ORFeome

Claims

What the coverage asserts, and how well corroborated each claim is across sources.

[● 3 SOURCES] The researchers excluded sequences capable of infecting humans, animals, plants, and fungi from the training data. (Stanford University and Arc Institute researchers)
[● 8 SOURCES] The ability to design complete viral genomes with AI raises significant biosafety and biosecurity concerns regarding potential misuse. (Stanford University and Arc Institute researchers)
[● 9 SOURCES] The synthetic viruses were designed to target Escherichia coli (E. coli) bacteria. (Stanford University and Arc Institute researchers)
[● 5 SOURCES] The AI generated approximately 700,000 potential viral genome designs. (Stanford University and Arc Institute researchers)
[○ 1 SOURCE] The Gates Foundation provided funding for the research led by Brian L. Hie. (Stanford University and Arc Institute researchers)
[● 11 SOURCES] The AI models used for the research are named Evo 1 and Evo 2. (Stanford University and Arc Institute researchers)
[● 11 SOURCES] Researchers at Stanford University and the Arc Institute used generative AI to design complete viral genomes. (Stanford University and Arc Institute researchers)
[● 11 SOURCES] Out of the designs tested, 16 synthetic viruses were found to be functional. (Stanford University and Arc Institute researchers)