A new machine learning system can predict the structure formed when two proteins dock, in a process that’s between 50 to 800 times faster than some software-based methods. This could help scientists better understand biological processes or speed the development of new therapies.
MIT researchers have developed a deep learning model that can rapidly predict the likely 3D shapes of a molecule given a 2D graph of its structure. This technique could accelerate drug discovery by narrowing down the number of molecules pharmaceutical researchers need to test in lab experiments.
In their quest to discover effective new medicines, scientists search for drug-like molecules that can attach to disease-causing proteins and change their