Stanford Team Uses Evo1 and Evo2 to Generate Complete Phage Genomes, With 16 of 302 Designs Killing E. coli

ITHome reported on August 7 that the BBC published an article today (August 7) stating that researchers at Stanford University in the United States had used AI technology to design novel, reproducible, fully functional viruses in a laboratory environment. The researchers said this marked the first time AI had successfully designed complete genomes. The 16 new viruses produced in the study infect bacteria and pose no threat to humans. In an interview with the BBC, Stanford University Assistant Professor Brian Hie said that AI had previously been used to design new antibiotics, but from a technical standpoint, designing a viable virus from scratch is more difficult. The research team said its approach works similarly to large language models such as ChatGPT, except that while large language models predict text sequences, the Evo 1 and Evo 2 models developed and used by the team predict the “language of life.” Regarding model training, the researchers said the data came from the genetic code of viruses, bacteria, plants, and humans. The team then fine-tuned the models to design bacteriophages, which are viruses that infect only specific bacterial species. ITHome note: In the context of large language models, fine-tuning refers to further training a pretrained large language model using a task-specific dataset. The pretrained model has already learned a large amount of general information, while fine-tuning helps it specialize in a particular field. To verify whether the AI-generated designs were effective, Stanford researchers selected the 302 most promising designs and synthesized them in the laboratory. The results showed that 16 of the designs could effectively kill E. coli. Developing new bacteriophages could provide a new treatment pathway for antibiotic-resistant infections. The report noted that phage therapy is regarded as a potential solution to the growing prevalence of bacterial infections, especially those that cannot be treated with antibiotics. The breakthrough also demonstrates that AI is now capable of designing novel biological systems that do not exist in nature. Hie believes this capability could significantly improve human health through the development of new drugs and therapies. The genetic code of the bacteriophages in this study was approximately 5,400 base pairs long. By comparison, the smallest living cell genome contains about 500,000 base pairs, while the human genome contains 3,000,000,000 base pairs. Hie said that attempting to design simple organisms “may require a great deal of work, but it is not impossible.” He added that the team is “indeed interested in working toward that goal.” Marc Güell, a professor at the Synthetic Biology Laboratory of Pompeu Fabra University in Spain, described the research as a “very important turning point” because “for the first time in history, humans are beginning to design biology on computers.” He said this makes it possible to envision more ways of addressing major challenges facing humanity, including developing bacteriophages to treat diseases, enzymes to treat genetic disorders, and antibodies for use in immunotherapy. Reference: Generative design of bacteriophages with genome language models