Researchers have successfully used artificial intelligence models to generate functional genetic code for a bacteriophage, a type of virus that infects and destroys bacteria. The AI-designed viruses demonstrated viability and, in some instances, proved more effective at targeting bacteria than their natural counterparts.
The project involved two distinct AI models tasked with creating complete genomes for a viable bacteriophage. As a reference, the researchers provided the models with the genetic sequence of ΦX174, a known bacteriophage that infects *E. coli* bacteria.
From approximately 700,000 potential designs generated by the AI, researchers selected 285 that appeared most promising. These designs were then synthesized into new DNA molecules. These synthetic DNA molecules were subsequently introduced into *E. coli* bacteria in Petri dishes.
Following this introduction, 16 of the Petri dishes exhibited clear signs of viral activity, indicated by the formation of plaques as the newly generated viruses began to infect and replicate within the *E. coli*. This confirmed the viability of the AI-designed bacteriophages. Notably, some of these synthetic viruses displayed enhanced efficacy in attacking *E. coli* compared to the original ΦX174 bacteriophage.
