Artificial intelligence could unlock bacteria-fighting viruses

Researchers say artificial intelligence could help scientists understand how viruses that infect bacteria might be harnessed as treatments for antibiotic-resistant infections.

Oleh Sama News Agency
30 Agustus 2026
A digital illustration showing an interconnected network of glowing white lines and nodes against a dark blue background, with bokeh light effects creating circular halos throughout.
An abstract visualization of interconnected networks and data, representing artificial intelligence and computational analysis that could help researchers study bacteriophages for treating antibiotic-resistant infections. (Phys.org)
2 menit membaca
Ukuran teks

Bacteriophages, viruses that infect bacteria, could one day help treat infections resistant to antibiotics, but scientists must first understand how these viruses work. Artificial intelligence may provide a powerful tool for that purpose, researchers say.

Known as phages, these viruses use bacteria to replicate themselves. They are thought to be more abundant on Earth than any other biological entity, having spent billions of years evolving ways to find bacteria, invade them and overcome their defenses. This diversity could benefit medicine, though finding the right phage for a given infection is not straightforward.

In a study published in the journal Science, Stanford University researchers used AI to design the DNA of a complete phage. When the synthetic DNA was constructed in the laboratory, it produced a functioning virus, demonstrating that AI can learn enough about a biological system to design something that works in the real world.

The Stanford team worked with ΦX174, a small, simple phage that infects E. coli. While it served as a good test case for demonstrating an AI-designed genome could be built and function, therapeutic phages of interest are five to 50 times larger and contain hundreds of genes with sophisticated molecular machinery allowing them to recognize bacteria, reproduce inside them and overcome bacterial defenses.

Much of phages' diversity remains unexplored, where AI could prove particularly useful. Software such as AlphaFold can predict the 3D structure of proteins from amino acid sequences, valuable for understanding phage proteins whose functions are unknown. Researchers at the Becky Mayer Centre for Phage Research at the University of Leicester and elsewhere isolate and study phages found in nature, repeatedly discovering unexpected biology. AI might identify genetic features determining which bacteria a phage can infect, reveal proteins that defeat bacterial defenses, or explain why some phages work better with particular antibiotics, researchers say. Every prediction would still require testing in the real world.

Artificial intelligence could unlock bacteria-fighting viruses | Sama News Agency