A drug that destroys one of the world's deadliest bugs has been discovered. Artificial intelligence (AI) could hold the key to combating antimicrobial resistance - named among the biggest threats facing mankind.
Acinetobacter baumannii is a hospital-acquired infection. It takes advantage of those with compromised immune systems, with premature babies being particularly vulnerable.
Canadian and US scientists used state of the art 'deep learning' to identify the antibiotic that kills it. Named Aubacin, it was effective in experiments on mice with infected wounds and bacterial cells grown in the lab.
The breakthrough was achieved using a neural network based on the human brain. It can access hundreds of millions - or even billions - of molecules with antibacterial properties.
The technique opens the door to finding treatments for other potentially fatal diseases - including MRSA and C diff. A. baumannii is notoriously difficult to cure. It can aggravate wounds and cause pneumonia, sepsis and meningitis - all of which are potentially fatal.
The bug can survive on surfaces such as door handles, cupboards and beds for long periods. It is able to pick up DNA from other species of bacteria in its environment - including antibiotic-resistance genes.
The researchers used an AI algorithm to predict the compound that can beat it. Conventional molecule screening techniques are challenging. They are time-consuming, costly and limited in scope.
Subscribe here for the latest news where you live
Lead author Dr Jonathan Stokes, of McMaster University, Hamilton, Ontario, said: "This work validates the benefits of machine learning in the search for new antibiotics. Using AI, we can rapidly explore vast regions of chemical space, significantly increasing the chances of discovering fundamentally new antibacterial molecules."
A. baumannii is on the World Health Organisation's (WHO's) list of 'priority pathogens' that urgently need new treatments due to the threat they pose.
Co author Professor James Collins, of Massachusetts Institute of Technology (MIT), said: "AI approaches to drug discovery are here to stay and will continue to be refined. 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 study, published in the journal Nature Chemical Biology, describes Abaucin as "especially promising" because it only targets A. baumannii. Crucially, it is less likely to rapidly develop drug resistance, offering hope of more precise and effective treatments.
Most antibiotics are broad spectrum in nature, meaning they kill all bacteria, disrupting the gut microbiome, which opens the door to a host of serious infections, including C diff. Added Dr Stokes: "We know broad-spectrum antibiotics are suboptimal and that pathogens have the ability to evolve and adjust to every trick we throw at them.
"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."
The WHO has declared antimicrobial resistance one of the top ten global public health threats against humanity. By 2050, it could claim ten million lives a year, costing the global economy around £100 trillion.