HAMILTON, Ontario — Artificial intelligence (AI) has helped scientists discover a new antibiotic capable of eliminating one of the world’s most deadly pathogens. This research breakthrough, enabled by advanced “deep learning” technologies, provides a potential solution to the pressing global challenge of antimicrobial resistance.
Acinetobacter baumannii, a hospital-acquired infection, often exploits patients with weakened immune systems, with premature babies being especially susceptible. The bacterium is notoriously challenging to treat and can survive on various surfaces, from door handles to beds, for extended periods. It’s known to acquire DNA, including antibiotic-resistance genes, from other bacterial species in its environment, causing severe complications such as pneumonia, sepsis, meningitis, and wound infections.
The antibiotic Aubacin, identified through a neural network model akin to the human brain, demonstrated its effectiveness against this deadly bacterium in lab experiments with mice and bacterial cells. This AI-driven approach can scan hundreds of millions, if not billions, of molecules with potential antibacterial properties, paving the way for new treatments for other lethal diseases, including MRSA and C. diff.
Conventional molecule screening techniques, which are laborious, expensive, and restricted in scope, are vastly outpaced by this AI-facilitated process.
“Using AI, we can rapidly explore vast regions of chemical space, significantly increasing the chances of discovering fundamentally new antibacterial molecules,” says Dr. Jonathan Stokes, the lead author of the study and a researcher at McMaster University, in a media release.
Scroll down to see 3 ways people can encounter antibiotic-resistant bacteria
The World Health Organization (WHO) has placed A. baumannii on its list of “priority pathogens” that urgently require new treatments due to their risk.
“AI approaches to drug discovery are here to stay and will continue to be refined,” says co-author professor James Collins, of Massachusetts Institute of Technology (MIT). “We know algorithmic models work. Now it is a matter of widely adopting these methods to discover new antibiotics more efficiently and less expensively.”
The unique aspect of Aubacin, described as “especially promising” lies in its selective target: A. baumannii. Unlike most antibiotics, which are broad-spectrum and can disrupt the gut microbiome by killing all bacteria, Aubacin’s specificity reduces the likelihood of rapid drug resistance development, opening a new chapter in the creation of more precise and effective treatments.
“AI methods afford us the opportunity to vastly increase the rate at which we discover new antibiotics, and we can do it at a reduced cost. This is an important avenue of exploration for new antibiotic drugs,” adds Dr. Stokes.
The WHO has identified antimicrobial resistance as one of the top ten global public health threats facing humanity. Predictions suggest that by 2050, it could result in ten million deaths annually, costing the global economy approximately $134 trillion.
The study is published in the journal Nature Chemical Biology.
3 situations people can encounter harmful superbugs:
- Hospitals: Despite being a place people go to get healthy, studies show antibiotic-resistant superbugs often contaminate hospital rooms, beds, and floors. A new study finds that these deadly bacteria can even make its way into the ICU. Researchers say an ICU in Hangzhou, China was entirely contaminated with a virulent strain that infected one in three patients — creating a nightmare for healthcare professionals.
- Meat: One of the strains people are most familiar with is E. coli, and meat products are a common source of these infections. Recently, researchers in Spain say they discovered multidrug-resistant E. coli strains in 40 percent of chicken, turkey, beef, and pork products for sale in shops across the country.
- Pets: Dogs and cats may be passing antibiotic-resistant “superbugs” to their owners, a new study warns. Researchers discovered one pet in the United Kingdom and six from Portugal were carrying similar antibiotic-resistant bacteria as their owners. These could include E. coli and other strains linked to pneumonia.
South West News Service writer Mark Waghorn contributed to this report.