AI-guided design of complete bacteriophage genomes

Summary

Artificial intelligence (AI) is increasingly being applied to the design of simple genetic systems, but its ability to generate whole functional genomes has remained untested. Writing in Science, King and colleagues report the generative design of complete bacteriophage genomes. The authors used the genomic language models Evo 1 and Evo 2, which were pretrained on large datasets of DNA sequences, including more than 2 million bacteriophage genomes. Bioinformatic benchmarking revealed that the AI models generated realistic sequences with diversity beyond that observed in nature. The authors then selected the lytic Microviridae bacteriophage ΦX174, which contains a genome of approximately 5.4 kilobases and 11 genes, as a design template. Using supervised fine-tuning of the models on approximately 15,000 Microviridae sequences, the authors generated ΦX174-like genomes, which were then computationally filtered on the basis of sequence quality, tropism specificity and evolutionary diversity, resulting in a curated list of 302 candidate bacteriophage genomes. This is a preview of subscription content, access via your institution Access options Access Nature and 54 other Nature Portfolio journals Get Nature+, our best-value online-access subscription 27,99 € / 30 days cancel any time Subscribe to this journal Receive 12 print issues and online access 251,40 € per year only 20,95 € per issue Buy this article - Purchase on SpringerLink - Instant access to the full article PDF. 39,95 € Prices may be subject to local taxes which are calculated during checkout Rights and permissions About this article Cite this article Ummarino, D. AI-guided design of complete bacteriophage genomes. Nat Biotechnol 44, 1448 (2026). https://doi.org/10.1038/s41587-026-03311-0 - Published: - Version of record: - Issue date: - DOI: https://doi.org/10.1038/s41587-026-03311-0