Wow! This sifting AI can do, in conjunction with bench top experimentation makes for a whole new game in drug discovery. From approx 39,000 substances for training the model, to 12 million compounds screened, to reveal 280 likely compounds, then off to the lab, and they found 2 promising candidates. Talk about finding a (2) needles in a hundred acres of haystacks!!

Citation: Discovery of a structural class of antibiotics with explainable deep learning (https://www.nature.com/articles/s41586-023-06887-8)