Claude Identifies CRISPR-Like Enzyme System in Bacteriophage DNA
Anthropic’s AI system found a previously uncharacterized structure called ART, but its biological function and gene-editing potential remain unknown.
Claude identified a previously uncharacterized enzyme system in bacteriophage DNA, finding a repeat structure resembling the arrays used by CRISPR systems and opening a new area for biological research.
The system, named array-associated reverse transcriptases, or ART, includes a reverse transcriptase, a neighboring partner gene and a long array of evenly spaced DNA repeats. Its biological function remains unknown, and further experiments are underway.
The comparison is based on the repeat-array structure and related characteristics. ART has not been shown to cut, copy or paste DNA, and no gene-editing application has been established.
Over a 21-hour search, about 950 Claude agents used 210 million tokens to examine genomic datasets for reverse transcriptase sequences. The agents collected more than 200,000 reverse transcriptases, identified 3,500 candidate systems and narrowed the results to 20 candidates for detailed reports.
One agent detected a repeating DNA pattern next to an unusual reverse-transcriptase gene. The pattern revealed a previously uncharacterized system in bacteriophages, viruses that infect bacteria.
The reverse transcriptase had been identified in earlier studies of a jumbo phage. Claude identified the associated repeat array and an additional partner protein as parts of a larger system.
Initial laboratory experiments found that the ART repeat array is expressed as distinct short RNAs. Human scientists supplied the initial prompt and conducted the laboratory work, while Claude agents performed the sequence analysis, evaluated candidates and proposed the discovery.
Anthropic established its life sciences research team and laboratory in spring 2026. The company’s Bay Area facility operates at biosafety levels BSL-1 and BSL-2 and does not handle pathogens capable of infecting humans.
“This is an exciting example of how AI agents can contribute to biological discovery,” Feng Zhang, a professor at MIT and the Broad Institute, said after reviewing the preprint.
Zhang called the identification of RNA-repeat arrays associated with reverse transcriptases “genuinely intriguing” and said it merits further investigation. Anthropic released a technical report as a preprint alongside the announcement.
