Deep Learning Approach for Discovering New Antibiotics

Discovery of a structural class of antibiotics with explainable deep learning - Nature 🔗

An explainable deep learning model using a chemical substructure-based approach for the exploration of chemical compound libraries identified structural classes of compounds with antibiotic activity and low toxicity.

The text discusses the urgent need for new structural classes of antibiotics to combat antibiotic resistance. It introduces a deep learning approach to explore chemical spaces and identify substructure-based rationales for compounds with high predicted antibiotic activity and low predicted cytotoxicity. The study empirically tests 283 compounds and identifies a structural class of antibiotics that is effective against antibiotic-resistant bacteria and reduces bacterial titres in mouse models. The approach enables the discovery of explainable structural classes of antibiotics and provides insights into the chemical substructures underlying selective antibiotic activity.

Would you like a bullet-point list summarizing the key points of the text?