Researchers at the Arc Institute and Stanford University used artificial intelligence models to design new viruses that infected and reproduced inside bacteria, according to a study published in Science and cited by The New York Times.
The team used genome language models called Evo 1 and Evo 2, which were trained on genetic sequences rather than text. The models were trained on trillions of nucleotides to learn patterns in genetic code.
For the experiment, the researchers further trained the models on about 15,000 viruses from the same family as Phi X-174, a virus that infects E. coli bacteria. They excluded viruses that could infect humans, other animals, plants or fungi.
Evo generated about 700,000 possible new viruses. The researchers selected 285 candidates, manufactured their DNA and inserted it into bacteria. Sixteen of the engineered viruses produced working viruses. The study said they were at least as viable as Phi X-174, and some reproduced faster.
The work points to possible uses in gene therapy, genome research and tools for attacking harmful bacteria. The source material also said the advance has raised concern that similar systems could be misused to design biological weapons.
The study was published in Science. The report cited by The New York Times said the researchers took precautions by limiting the work to viruses that infect bacteria.
Featured image credit
