Researchers at the University of Oregon have developed an artificial intelligence tool that can read genetic code the way large language models like ChatGPT read text. Scanning the genome for biological mutation patterns, the computer model traces the ancestry of genes across different species, according to a study published in Nature Communications on March 18, 2026.
The AI model, called GenoFormer, was trained on genomic data from over 100,000 species. It identifies patterns of mutations that have accumulated over evolutionary time, allowing it to reconstruct gene family trees without prior knowledge of species relationships. Lead author Dr. Emily Chen stated, 'This is like having a universal translator for the language of DNA.'
In tests, GenoFormer accurately traced the ancestry of genes involved in immune response and metabolism, matching known evolutionary trees built through traditional methods. The tool also discovered previously unknown relationships between genes in bacteria and plants, suggesting ancient horizontal gene transfer events.
The University of Oregon team plans to make the model publicly available for other researchers. Potential applications include tracking the evolution of antibiotic resistance genes and understanding how diseases emerge across species. The work was funded by the National Science Foundation.
