Scientists at the University of Pennsylvania have utilized an AI algorithm to predict nearly one million new antibiotic molecules from the global microbiome, according to a study published in the journal Cell on June 5, 2024.
The research team, directed by Professor César de la Fuente of the Machine Biology Group, developed an algorithm to mine vast amounts of microbial data, expediting antibiotic discovery processes.
This groundbreaking AI tool identified DNA snippets with potential antimicrobial activity, which were then synthetically reproduced and tested. Of the 100 synthesized molecules, 79% demonstrated the ability to kill one or more microbes, suggesting potential as antibiotics. The team made their data and code publicly accessible to encourage further exploration and validation by other researchers.
The urgency of this research is highlighted by the rising threat of antimicrobial resistance, responsible for over 1.2 million deaths in 2019, with projections by WHO suggesting this could escalate to 10 million deaths annually by 2050. The study signifies a major advancement in antibiotic resistance research powered by artificial intelligence, potentially cutting down the typically lengthy drug discovery timeline.