Study shows AI can design bacteriophages with potential to overcome bacterial resistance
Scientists used genome language models to design complete bacteriophage genomes, with laboratory tests showing that some could infect E. coli and overcome resistance to a naturally occurring phage
A study published in Science reveals that artificial intelligence can be employed to design entire bacteriophage genomes, potentially offering new alternatives to antibiotics for combating drug-resistant bacterial infections. Scientists from Stanford University and the Arc Institute utilized AI models named Evo 1 and Evo 2 to generate thousands of potential bacteriophage genomes from a single framework, the naturally occurring bacteriophage ΦX174.
Out of nearly 300 chemically synthesized designs, sixteen produced functional viruses capable of infecting Escherichia coli. The AI-generated phages exhibited novel combinations of genes and regulatory elements, differing from existing viruses in genome length and containing DNA-packaging proteins evolutionarily distant from those normally associated with their capsids.
Notably, a cocktail of these newly designed phages rapidly overcame resistance in laboratory E. coli strains previously resistant to ΦX174, suggesting AI could expand the range of phage designs. However, the study does not address human infections, and extensive further testing is needed to assess their safety and effectiveness as therapies.
Experts warn that such generative AI technologies raise biosafety and biosecurity concerns, as similar tools could potentially be applied to human, animal, or plant pathogens with unpredictable biological behavior.
Written by urgent.news from The Hindu - Sci-Tech's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
