Stanford researchers use AI to design functional bacteriophage genomes

Researchers from Stanford University and the Arc Institute have successfully used artificial intelligence to design complete bacteriophage genomes. Using genomic language models known as Evo 1 and Evo 2, the team trained the AI on vast amounts of DNA sequences to predict nucleotide patterns, similar to how large language models process human text.

While the AI generated thousands of potential genomes, approximately 5% were successfully synthesized in a laboratory and proved to be functional viruses capable of infecting and reproducing within Escherichia coli. This breakthrough demonstrates that AI can learn the underlying rules of genomic organization to propose new, viable biological sequences, offering significant potential for advancements in synthetic biology and biotechnology.