With nearly 5 million deaths linked to antibiotic resistance globally every year, the urgent need for new ways to combat resistant bacterial strains is undeniable. Researchers at Stanford Medicine and McMaster University are tackling this challenge head-on, leveraging the power of generative artificial intelligence with a novel model that has successfully created structures and chemical recipes for six new drugs aimed at killing resistant strains of Acinetobacter baumannii, one of the leading pathogens responsible for antibiotic resistance-related deaths.
The team detailed their findings in a study published in Nature Machine Intelligence."There's a huge public health need to develop new antibiotics quickly," said James Zou, co-senior author on the study. "Our hypothesis was that there are a lot of potential molecules out there that could be effective drugs, but we haven't made or tested them yet. That's why we wanted to use AI to design entirely new molecules that have never been seen in nature."
Unlike previous computational approaches that sifted through existing drug libraries, SyntheMol was trained to construct potential drugs using a library of over 130,000 molecular building blocks and a set of validated chemical reactions. This allowed the model to not only generate new compounds but also provide the steps to produce them in a lab setting.
The researchers' efforts paid off, as SyntheMol generated around 25,000 possible antibiotics in less than nine hours. Six of these newly generated compounds were found to be effective against the resistant strain of A. baumannii, as well as other infectious bacteria prone to antibiotic resistance, such as E. coli, Klebsiella pneumoniae, and MRSA.
"This AI is really designing and teaching us about this entirely new part of the chemical space that humans just haven't explored before," Zou added.