Artificial intelligence accelerates discovery of next-generation disinfectants

Chemists and computer scientists collaborated to use AI to discover new disinfectants to fight antibiotic-resistant “superbugs”.
The team developed a computational-experimental framework to design quaternary ammonium compounds (QACs), achieving 11 new QACs effective against resistant bacteria.
This work marks the first example of using AI to generate molecules specifically for disinfectants.
The method included creating a large dataset of existing QAC molecules, feeding this data into a machine learning model, and then synthesizing and testing the AI-generated compounds.
The AI-driven process significantly improved when the model was retrained with only the most potent QAC molecules, reducing invalid compounds and boosting the number of promising new candidates from 9% to 38%.
The new, broad-spectrum QACs are now being studied further, and the framework has potential applications to speed discovery of other effective disinfectants.
The project serves as a model for integrating AI with experimental chemistry, especially for fields lacking extensive data.

You may have missed