A chemical ‘reactome’ platform has been created by researchers from the University of Cambridge and Pfizer, using automated experiments and AI to predict how chemicals will react with one another. This method, applied to pharmaceutically relevant reactions, can speed up the drug design process by revealing hidden relationships between reaction components and outcomes. Additionally, the researchers have developed a machine learning model for storing data on the precise core variations of molecules, overcoming the challenge of relatively few reported late-stage reactions, by predicting the sites of reactivity, that is the likely locations of atoms reacting, under different conditions.
The whytry.ai article you just read is a brief synopsis; the original article can be found here: Read the Full Article…
