AI Is Learning to Write Genetic Code
Researchers have successfully used AI to design and synthesize entirely new bacteriophages – viruses that target bacteria – demonstrating a significant advancement in synthetic biology and raising concerns about potential misuse. The AI models generated viable viruses that outperformed existing ones, highlighting both the promise and the potential dangers of this technology.
Researchers have achieved a remarkable feat in synthetic biology by leveraging artificial intelligence to design and create novel bacteriophages. The project centered around two AI models tasked with generating complete genomes for viable bacteriophages, using the existing ΦX174 bacteriophage as a reference point. The models produced approximately 700,000 potential genome designs, from which the researchers selected 285 that appeared most promising based on their predicted viability. These selected designs were then used to synthesize new DNA molecules, which were subsequently inserted into *E. coli* bacteria. Within a short period, 16 of the *E. coli* cultures began to exhibit clear signs of viral infection and replication, confirming the viability of the newly synthesized viruses. Notably, some of these new viruses demonstrated superior effectiveness against *E. coli* compared to the original ΦX174 bacteriophage, showcasing the AI's ability to generate effective biological agents.