According to WIRED, researchers have used an artificial intelligence system for the first time to design a series of previously unknown viruses capable of infecting and eliminating certain types of bacteria. Published this week in the journal Science, the work opens new possibilities for combating bacterial resistance, but it also raises concerns about the potential misuse of the technology to design biological weapons, WIRED reported.
From natural genomes to new pathogens
Scientists have been able to synthesize viruses from scratch for several years, typically using the technique to develop and evaluate antiviral drugs and vaccines and to expand understanding of how these microorganisms behave, according to WIRED. That process, however, has relied primarily on replicating previously known pathogens or their variants.
In the new study, scientists at Stanford University and the Arc Institute used AI to design simple, functional, and previously unseen viruses based on information contained in the genetic sequences of millions of animals, plants, microbes, bacteria, and viruses found in nature, WIRED reported. The researchers worked with bacteriophages — viruses with relatively small genomes that infect only bacteria, making them an appealing tool for biotechnology and a potential alternative to antibiotics against resistant infections.
The creation of the new viruses was based on Evo 1 and Evo 2, foundational AI models developed for computational biology. Both algorithms were trained on millions of genomes from all domains of life, with the goal of learning complex evolutionary patterns, including how genes are organized, which sequences are conserved, and the biological constraints that allow an organism to remain functional.
The experimental design used the bacteriophage Phi X-174 — which infects Escherichia coli (E. coli) — as a reference. The objective was not to reproduce this virus, but to use it as a guide for the algorithms to generate thousands of new genomes with a genetic architecture compatible with infecting E. coli. The resulting viruses retained the functional organization needed to recognize the bacterium, insert DNA, replicate, produce new particles, and assemble them, yet their specific DNA sequences differed considerably from naturally occurring bacteriophages.
300 genomes, 16 functional viruses
The scientists evaluated the AI-generated genomes, selecting those most likely to be functional based on factors such as gene organization, the presence of regulatory elements, and other criteria inspired by the biology of Phi X-174. This selection produced a sample of 300 genomes, which were synthesized molecule by molecule in the laboratory, then introduced into E. coli bacteria to test whether they produced functional viruses.
| Step | Result |
|---|---|
| AI-generated genomes evaluated | 300 selected as most likely functional |
| Genomes synthesized in lab | All 300 |
| Introduced into E. coli bacteria | 16 produced fully functional bacteriophages |
| Characteristics of the 16 | Previously unpublished sequences, different genes, new regulatory elements, varying genome sizes |
| Behavior of the 16 | Some infected bacteria more quickly; others replicated differently |
According to WIRED, only 16 of the 300 synthesized genomes gave rise to fully functional bacteriophages, with previously unpublished sequences, different genes, new regulatory elements, and even varying genome sizes. The viruses also behaved differently from one another: some infected bacteria more quickly, while others exhibited different replication abilities.
AI-designed phages beat resistant bacteria
The research also evaluated the ability of the AI-generated bacteriophages to combat resistant bacteria. WIRED reported that the experiment exposed a mixture of AI-designed phages and a mixture of natural phages similar to Phi X-174 to strains of E. coli that had already developed resistance to that virus. The results showed that the AI-generated viruses were able to rapidly overcome bacterial resistance and establish infection. According to the authors, the finding demonstrates a path toward a new approach to resistant infections.
Implications and concerns
The breakthrough, WIRED reported, opens up new possibilities for combating bacterial resistance, while also raising concerns about the potential misuse of the technology to design biological weapons. The study demonstrates that AI can generate functional viruses without replicating previously known pathogens, though only 16 of the 300 synthesized genomes produced working bacteriophages.
Sources:
