Scientists at Stanford University have successfully employed artificial intelligence to design fully functional viral genomes for the first time, according to research published in the journal Science. The effort resulted in 16 novel bacteriophages—viruses that infect bacteria—which were synthesized in the laboratory and demonstrated the ability to replicate and kill E. coli bacteria.
The AI systems used, called Evo1 and Evo2, operate on principles similar to large language models such as ChatGPT, but predict genetic sequences rather than text patterns. The models were trained on genetic information from viruses, bacteria, plants, and humans, then refined specifically to generate bacteriophage designs. From approximately 302 candidate designs generated by the AI, researchers synthesized and tested those showing the most promise, ultimately confirming that 16 performed effectively in laboratory conditions.
The breakthrough represents a significant advancement in synthetic biology and has potential applications in treating antibiotic-resistant infections. Experts note that bacteriophage therapy could provide new approaches to disease treatment as traditional antibiotics face growing resistance challenges. Additionally, researchers suggest the technology could facilitate development of novel drugs and therapeutic approaches addressing various human health challenges.
However, the achievement has prompted warnings from biosecurity experts regarding potential misuse. Commentary in Science from researchers at Johns Hopkins University’s Center for Health Security highlighted “urgent” biosafety and biosecurity questions, noting that the ability to design viral genomes raises concerns about potential malicious applications. The Stanford team implemented multiple safety measures, including excluding viruses capable of infecting complex organisms from training data and conducting all work in secure laboratory settings.
Researchers acknowledge significant technical barriers remain before AI could design more complex living organisms, given the vast difference in genome size between the 5,400 base-pair phages and the smallest cellular genomes at approximately 500,000 base pairs. Nevertheless, scientists expressed interest in pursuing more complex organisms in future research, while advocating that existing safeguards provide sufficient protection for responsible development of the technology.
Article Attribution | Read More at Article Source
Article summary produced by Claude AI