Artificial intelligence in the fight against multi-resistant germs

Hardly any other germ is feared more in hospitals than Acinetobacter baumannii, known as the "multidrug-resistant germ." It is classified by the World Health Organization as one of the world's most dangerous antibiotic-resistant bacteria. Over time, these strains of bacteria have developed high levels of resistance to a variety of antibiotics and are very difficult to treat for this reason. They are particularly widespread in medical facilities, where they can cause life-threatening infections. Thanks to the use of artificial intelligence, medical research has now succeeded in finding an antibiotic molecule against this invisible enemy: abaucin.

The groundbreaking discovery was made by researchers at McMaster University and the Massachusetts Institute of Technology. They used a specially developed algorithm that screened about 7500 molecules for their antibacterial properties. To achieve this, the researchers trained a neural network with data from a test of Acinetobacter baumannii growth inhibition and used artificial intelligence to make accurate predictions about structurally novel molecules. The result was a molecule called abaucin, which has already been successfully used to fight infection in mice.

This promising result raises hope in the fight against multidrug-resistant germs. But it also lays the groundwork for the use of artificial intelligence in medical research, as the study's lead author and assistant professor in McMaster University's Department of Biomedical Sciences and Biochemistry Jonathan Stokes also points out: "This work confirms the benefits of machine learning in the search for new antibiotics. We know that broad-spectrum antibiotics are suboptimal and that pathogens have the ability to adapt and evolve to any method we use. AI methods allow us to significantly increase the rate at which we discover new antibiotics, at a lower cost. This is an important discovery pathway for new antibiotics."