Stanford Researchers Use Large Genome Models to Design New Viruses
Researchers at Stanford University have leveraged large genome models, a type of AI trained on vast amounts of DNA sequence data, to design new viruses that infect bacteria. This work extends previous applications of such models, which had already been used to generate DNA sequences encoding functional proteins and mimicking gene structures in complex cells. The newly designed viruses are closely related to an existing virus, but they possess distinct features that would be challenging to evolve naturally. The researchers suggest that society should begin considering the potential misuse of similar AI technologies to design viruses targeting vertebrates. This development highlights the growing capabilities of AI in biology, moving beyond protein design to whole-genome engineering, and raises important ethical and biosecurity concerns.
Key facts
- Large genome models have been used to design new viruses that infect bacteria.
- The research was conducted at Stanford University.
- The designed viruses are closely related to an existing virus but have distinct features.
- The viruses have features that would be challenging to evolve naturally.
- Previous work with large genome models produced DNA sequences encoding functional proteins in bacteria.
- Large genome models have also mimicked gene structures found in complex cells.
- The researchers suggest preparing for potential AI-designed viruses targeting vertebrates.
- The work is part of a broader trend of AI applications in biology, initially focused on protein design.
Entities
Institutions
- Stanford University