Testing-Robot for Drug-Protein Interactions
March 15, 2016 | Terry Sharrer
While molecular diagnostics lead the way in “patient discovery,” drug discovery proceeds with predicting interactions between drugs and their protein targets. To that end, researchers at Carnegie Mellon University have used a liquid handling robot, combinatorial chemistry techniques and machine learning to sort through the enormous number of possible interactions. In the example presented in this piece, their resulting algorithm was able to predict how 96 drugs affected 96 proteins with 92% accuracy, based on a small percent of possible experiments. MORE