antibiotics is it's easier to discover new antibiotics that have similar chemical structures and means of action to the ones that we already know. But those are also the ones that bacteria know, because they're similar to the ones that are out there. And so what you really want to do in principle in discovering antibiotics is to find compounds that are quite different from existing compounds. But that's also a place that's very hard for human beings to explore. And what this work on Hallison, a new compound that was discovered at MIT with AI, has demonstrated is the ability of a solution that involves both deep expertise. So this is not just machine learning experts, it's some of the best synthetic biology people in the world as well, but involves deep human expertise, but also AI to identify potential new compounds that are very far away, very different from the existing compounds that we have. And it's this ability to expand the nature of how it is that we discover things. It's not (15/45)
You are viewing a single comment's thread from: