A research team from Stanford University and the Arc Institute in California has developed artificial intelligence (AI) to create functional viruses that target bacteria, marking a significant advance in the generative design of complete genomes. According to a preprint paper, the AI system proposed new genetic codes for viruses, leading to the synthesis of 302 DNA strands that were tested against E. coli bacteria. Of these, 16 AI-generated viruses were successful in replicating and destroying their bacterial host.
The AI, named Evo, operates similarly to a large language model but is specifically trained on approximately two million bacteriophage genomes. It was tasked with generating variants of the phiX174 bacteriophage, which contains 11 genes and about 5,000 DNA letters. Researchers noted that the AI’s designs included innovative gene order changes not previously considered by scientists.
This technology may enhance phage therapy for multidrug-resistant bacterial infections and improve gene therapy methods. However, experts also caution against potential risks, particularly if similar methods are applied to dangerous viruses. Additionally, further complexity arises when attempting to scale this method to living cells, given the significantly larger DNA of bacteria compared to that of the tested viruses.
- Why it matters: This research could lead to more effective treatments for antibiotic-resistant infections and advance genetic therapies, impacting healthcare and biotechnology sectors.
- The latest: Concerns about the safety of AI-generated viruses have been raised, emphasizing the need for caution in future applications.
Source: https://the-decoder.com/stanford-and-arc-institute-scientists-used-ai-to-design-new-viruses-that-killed-bacteria-in-the-lab/
Source: https://the-decoder.com/stanford-and-arc-institute-scientists-used-ai-to-design-new-viruses-that-killed-bacteria-in-the-lab/