Researchers from the University of Toronto have developed a new set of machine learning algorithms that can generate 3D structures of tiny protein molecules. These algorithms could completely transform the development of drug therapies for numerous diseases, including cancer and Alzheimer’s.
Designing successful drugs is a complex process, sort of like solving a really complicated puzzle. And without knowing the 3D shape of a protein, it’s like trying to solve that puzzle without being able to look at it. But the ability to determine the three-dimensional shape of protein molecules is critical if the researchers want to know how they will respond to drug therapies.
The new set of algorithms is able to reconstruct 3D structures of protein molecules using microscopic images. Because proteins are so tiny, they can’t be seen without using techniques such as electron cryomicroscopy (cryo-EM). Cryo-EM uses high-power microscopes to take tens of thousands of low-resolution images of a frozen protein sample from different positions. However, the problem that remains is how to piece together the correct high-resolution 3D structure from the low-resolution 2D images.
This is where the new algorithms come in. “Existing techniques take several days or even weeks to generate a 3D structure on a cluster of computers,” said Marcus Brubaker, an Assistant Professor at York University. “Our approach can make it possible in minutes on a single computer.”
These algorithms could not only lead to new drug candidates for a range of diseases, they could also lead to a new understanding of how life works at the atomic level.
University of Toronto (https://www.utoronto.ca/news/new-algorithms-u-t-researchers-may-revolutionize-drug-discoveries)