AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Machine Learning for Drug Discovery
All the standard doors for for drug discovery were closed. You couldn't run a big physical screen, you couldn't design basinstructor. We used a very, very distant protein, treal protess, was separated by three billion years of evolution. Only 20 % of the menoacids were identical. So very, very distantly related. But we used as a homologe model. Out of seven million, we pull down to 60 thaugh we thought we'd test. And five of the 60 were actually accurate, with the best one more potent than what you would expect from a physical screen. That's so col it really is.