Hospital germ: AI finds effective antibiotic
With the help of artificial intelligence (AI), researchers at Massachusetts Institute of Technology (MIT) and McMaster University have identified a new antibiotic that could play an important role in the fight against hospital germs, as multi-resistant bacteria are often called.
Hospital germs are particularly dangerous because they are already resistant to many known antibiotics and are also able to develop new resistances quickly.
One of the most problematic is Acinetobacter baumannii, which can cause pneumonia, meningitis, and other serious illnesses. The perfidious thing about the dangerous bacterium: It can survive on hospital doorknobs and equipment for long periods of time and pick up antibiotic resistance genes from its environment, explained Jonathan Stokes, assistant professor of biochemistry and biomedical sciences at McMaster University, in one press release.
That is why it is now common to find traces of the bacterium “which are resistant to almost every antibiotic”.
To find a successful anti-Acinetobacter baumannii compound, the research team trained a machine-learning model they developed with lab results obtained by exposing the bacterium to around 7,500 different compounds.
From those that inhibited the growth of the bacterium, the model selected “a few hundred” candidates that were promising. 240 of these were then tested in the classic way with bacterial cultures, which the AI significantly reduced the effort, because without the model all substances would have had to be tested in this way.
This is how the scientists came across the substance that is very effective against Acinetobacter baumannii – and gave it the name Abaucin. Animal studies have already started, but the data is not yet sufficient, the researchers said.
For James Collins from MIT, however, this discovery is an unmistakable sign that we are on the right track: “This finding supports the assumption that AI can significantly accelerate and expand our search for novel antibiotics.”
