Pharmacy Daily | AI could help win the war against AMR

MACHINE learning has identified 863,498 promising antimicrobial peptides offering a potential breakthrough in the fight against antimicrobial resistance (AMR).

A team of researchers, including Queensland University of Technology (QUT) computational biologist A/Prof Luis Pedro Coelho (pictured), uncovered the small molecules that can kill or inhibit the growth of infectious microbes.

Published in the journal Cell, the study arrives at a crucial time as the world grapples with a rising number of superbugs resistant to existing drugs.

"There is an urgent need for new methods for antibiotic discovery," said Coelho, a researcher at the QUT Centre for Microbiome Research.

"AMR is a major public health threat, killing 1.27 million people annually, and without intervention, it could cause up to 10 million deaths per year by 2050.

"Using artificial intelligence to understand and harness the power of the global microbiome will hopefully drive innovative research for better public health outcomes," Prof Coelho added.

The research team tested 100 laboratory-made peptides against clinically significant pathogens, verifying machine predictions.

They found that 79 disrupted bacterial membranes, and 63 specifically targeted antibiotic-resistant bacteria such as staphylococcus aureus and escherichia coli.

Some peptides helped eliminate infections in mice, with two reducing bacteria by up to four orders of magnitude.

In preclinical models, treatment with these peptides produced results similar to polymyxin B, a commercially available antibiotic used to treat severe infections like meningitis, pneumonia, and sepsis.

The team analysed over 60,000 metagenomes, containing the genetic makeup of more than one million organisms from diverse environments including marine, soil, and human and animal guts.

The study's findings are in the AMPSphere database, a resource for new antibiotic discovery. JG