Generative design of novel bacteriophages with genome language models by Samuel H. King, Claudia L. Driscoll, David B. Li, Daniel Guo, Aditi T. Merchant, Garyk Brixi, Max E. Wilkinson, Brian L. Hie ISBN 10.1101/2025.09.12.675911 instant download

Generative design of novel bacteriophages with genome language models by Samuel H. King, Claudia L. Driscoll, David B. Li, Daniel Guo, Aditi T. Merchant, Garyk Brixi, Max E. Wilkinson, Brian L. Hie ISBN 10.1101/2025.09.12.675911 instant download

AbstractMany important biological functions arise not from single genes, but from complex interactions encodedby entire genomes. Genome language models have emerged as a promising strategy for designingbiological systems, but their ability to generate functional sequences at the scale of whole genomeshas remained untested. Here, we report the first generative design of viable bacteriophage genomes.We leveraged frontier genome language models, Evo 1 and Evo 2, to generate whole-genome sequenceswith realistic genetic architectures and desirable host tropism, using the lytic phage ΦX174 as our designtemplate. Experimental testing of AI-generated genomes yielded 16 viable phages with substantialevolutionary novelty. Cryo-electron microscopy revealed that one of the generated phages utilizes anevolutionarily distant DNA packaging protein within its capsid. Multiple phages demonstrate higherfitness than ΦX174 in growth competitions and in their lysis kinetics. A cocktail of the generated phagesrapidly overcomes ΦX174-resistance in three E. coli strains, demonstrating the potential utility of ourapproach for designing phage therapies against rapidly evolving bacterial pathogens. This work providesa blueprint for the design of diverse synthetic bacteriophages and, more broadly, lays a foundation forthe generative design of useful living systems at the genome scale.