science
Of course, it’s not about killing humans, but about killing the antibiotic-resistant bacteria that are causing their infection.
Researchers have given AI-designed bacteria-targeting viruses to a person, which they claim is a first, but aren’t rushing them to get results: they’re not yet ready to say how it happened.
Radical Numerics, a San Francisco-based AI lab developing AI models for biosciences, announced the treatment this week, saying it was the first time a cocktail containing AI-designed bacteriophages had been administered to a human patient under a compassionate use agreement.
Be careful, these aren’t just any viruses: Omnii-1, the AI model publicly unveiled by Radical Numerics this week, was used to design viable bacteriophages, a type of virus that specifically infects bacteria.
The therapeutic use of phages is not new in itself, with bacteria-targeting viruses being sought as an alternative to combat the rise of antimicrobial-resistant bacteria that increasingly render antibiotics ineffective.
Phages have a number of potential advantages over antibiotics, including their ability to target specific bacterial strains – a huge advantage when it comes to sparing beneficial gut bacteria that can be affected by broad-spectrum antibiotics.
AI has already shown promise for accelerating phage engineering, as we reported last year when a group of researchers from Stanford University and the Arc Institute created AI-designed phages targeting the E. coli bacteria. The phages, designed using the genetic sequence of a well-studied natural phage of E. coli, included several variants that outperformed their biological predecessor in laboratory tests.
Similar to Stanford’s work, referenced in the Radical Numerics article, the company used its own AI model to design bacteriophages, which it then tested against bacteria isolated from the antibiotic-resistant infection his patient suffered from.
“The cocktail containing Omnii-1 engineered phage variants eliminated the patient’s antibiotic-resistant isolate in vitro,” the company explained, so there is at least one success recorded there. “The experimental cocktail was then administered to the patient on a compassionate basis.”
Unfortunately, this is where the list of results ends, as Radical Numerics has not published anything about the results of testing the phage on a human, citing the need to protect privacy. The company also doesn’t want to give the impression that its AI phage therapy is generally safe or effective, noting that observations from a single patient don’t establish much.
It’s not even specified what infection the patient had, although the research report mentions testing Omnii-1’s ability to predict which phages might infect Pseudomonas aeruginosa, a bacteria that can develop resistance to antibiotics and cause infections in patients with burns and other wounds, as well as persistent lung infections in patients with cystic fibrosis. The AI model “performed competitively” against a specialized model in predicting phage-host interactions, the research report notes.
There is nothing in the document to suggest that the patient suffered from a Pseudomonas aeruginosa infection.
“This marks an important milestone in translating AI-designed biology, computational predictions and laboratory validation to clinical administration,” Radical Numerics said of its work. The treatment takes the technology beyond laboratory testing, although it fails to completely eliminate the infection. Radical Numerics hopes its work will lead to a process of designing and building phages targeted at individual bacterial infections, and the need for alternatives is already being felt: Antibiotic resistance poses a growing global health threat, with drug-resistant infections already causing significant mortality worldwide.
Radical Numerics did not respond to questions for this story. ®
Gn Health
