AI aids search for antibiotic to beat stubborn bacteria
Artificial intelligence (AI) has helped scientists to discover a new antibiotic that could potentially treat drug-resistant bacteria.
Scientists have used artificial intelligence (AI) to discover a new antibiotic that could potentially treat drug-resistant bacteria.
In a study published in the Nature Chemical Biology journal, scientists used an AI algorithm to sift through more than 7,000 drug compounds to find one that could tackle a bacterium called Acinetobacter baumannii.
The World Health Organisation has listed this bacterium as one of three pathogens that researchers should be breaking their backs to find an antibiotic for.
A 2008 study published in the American Society of Microbiology (ASM) journal described Acinetobacter baumannii as “a highly problematic pathogen for many institutions worldwide”.
It stated that its cunning nature makes it resistant to drugs and gives it the ability to stay longer wherever it finds a home—it thrives in hospitals.
“The organism often targets the most vulnerable hospital patients, those who are critically ill with breaches in skin integrity and airway protection. As reported in reviews dating back to the 1970s, hospital-acquired pneumonia is still the most common infection caused by this organism,” stated the ASM study.
Other studies have shown that the bacterium causes meningitis and has also infected wounded military personnel.
“Acinetobacter can survive on hospital doorknobs and equipment for long periods, and it can pick up antibiotic resistance genes from its environment. It’s really common now to find A. baumannii isolates that are resistant to almost every antibiotic,” explained Jonathan Stokes, lead author of the study, on a Massachusetts Institute of Technology website.
In the new study, the researchers highlight the urgency of finding a new antibiotic for this bacteria, which they say is “notoriously difficult to eradicate due to its ability to acquire and retain antibiotic resistance determinants”.
Although this study was conducted in mice, the results offer hope for finding a new antibiotic. The mice used in the study were infected with the bacteria and then treated with the antibacterial compound abaucin.
“The machine learning-guided discovery of abaucin highlights the utility of algorithmic approaches to discover new antibacterial molecules against A. baumannii and provides the field with a promising new narrow-spectrum molecular scaffold to combat one of the world’s most challenging Gram-negative pathogens,” the study notes.
The compound specifically targets A. baumannii, which means the pathogen is less likely to rapidly develop drug resistance and could lead to more precise and effective treatments.
Dr Stokes also said the work validates the benefits of machine learning in the search for new antibiotics.
“Using AI, we can rapidly explore vast regions of chemical space, greatly increasing the chances of discovering fundamentally new antibacterial molecules,” he said.
“We know that broad-spectrum antibiotics are suboptimal and that pathogens have the ability to evolve and adapt to any trick we throw at them,” Dr Stokes said.
He added: “AI methods give us the opportunity to significantly increase the rate at which we discover new antibiotics, and we can do it at a lower cost. This is an important avenue of discovery for new antibiotics.”
