Researchers from York University’s Lassonde School of Engineering have developed a new set of algorithms that rapidly generate 3D structures of proteins, and could revolutionize the development of new drug therapies.
One of the lead researchers on the project, Lassonde Professor Marcus Brubaker, says the current cryo-EM (electron cryomicroscopy) technology for developing 3D protein structures is a lengthy, computationally demanding task requiring specific expertise. Currently, it can take days to weeks for certain results.
Researchers from York University’s Lassonde School of Engineering have found a new set of algorithms that can help determine the 3D structure of proteins, which could one day lead to new treatments for diseases including Alzheimer’s, HIV and cancer
The algorithms researchers developed are combined in a software program called cryoSPARC (cryo-EM single-particle ab initio reconstruction and classification), which enables non-specialized cryo-EM users to process data in a matter of hours.
“Collecting data on an electron microscope might take a few hours or maybe a day or two,” said Brubaker. “However, processing that data to determine the 3D structure would require weeks or even months of computation time on large, expensive computer clusters. Our work now makes this possible in a few hours on a relatively inexpensive desktop computer.”
The research is published in the current edition of the journal Nature Methods.
The dramatic change in processing times not only speeds up the existing process, but also enables experts to dig deeper into their data to discover new biology that, before, would not have been practical. It also has potential for enhanced research into drug treatments for a range of diseases, including Alzheimer’s, HIV and cancer.
“Our developments have also enabled us to determine structures without any prior knowledge, opening up an entire new class of molecules that were unable to be studied otherwise,” said Brubaker.
Drugs work by changing properties of specific proteins in the body. For a drug to be successful, it must be designed with a specific shape so it binds only to the desired protein, as binding with other proteins could can cause side effects.
The algorithms, co-developed by U of T PhD student Ali Punjani, could significantly aid in the development of new drugs because they provide a faster, more efficient means of arriving at the correct structure.
“Any symptom or disease in our body has some protein interaction component to it,” said Brubaker. “So, whether it’s Alzheimer’s or cancer, our ability to understand what’s happening at the cellular level and then target those behaviours is really the basis of treatment and diagnosis of disease. To the extent that we’re able to develop tools to allow researchers to study these structures in ways they’ve never been able to before, the impact is boundless in terms of what it could mean for disease research.”
Together, Punjani and Brubaker founded the Toronto-based startup Structura Biotechnology Inc., which is developing the software package cryoSPARC for use in academic and industrial labs. Structura has received funding and support from U of T’s Innovations & Partnership’s Office (IPO) through the Connaught Innovation Award, U of T’s Early Stage Technologies (UTEST) program, the Ontario Centres of Excellence (OCE) and FedDev Ontario’s Investing in Commercialization Partnerships program at York University.
The research was done in collaboration with U of T Professors David Fleet and John Rubinstein, with funding from the Natural Sciences & Engineering Research Council of Canada (NSERC).