More and more, the fields of biology and genomics are becoming big data sciences. Today research discovery and innovation are made possible by not only experiments in the laboratory, but also through the power of high performance computing (HPC). Without access to such computational resources, it would not be possible to analyze and interpret the terabytes of data generated every day that contribute to scientific discovery.
Canadian health care organizations have been lacking the HPC infrastructure they need to push research and personalized clinical care into the future. To address this problem, The Hospital for Sick Children (SickKids) and University Health Network’s (UHN) Princess Margaret Cancer Centre have partnered to build a pilot IT infrastructure that will provide researchers and clinicians with secure cloud-computing services, that will also satisfy personal health information privacy requirements. The pilot project is called High Performance Computing for Health Sciences, also known as HPC4Health.
“Together, with Princess Margaret Cancer Centre we designed HPC4Health, a resource-sharing model to increase efficiencies not only to operating budgets, but also to help clinicians analyze patient tests faster,” says Dr. Michael Brudno, Senior Scientist and Director of the Centre for Computational Medicine at SickKids, and Associate Professor in the Department of Computer Science at University of Toronto.
HPC4Health will allow for faster scientific discovery, faster diagnosis for patients and faster identification of therapies for patients, not only in the field of genetics, but also in medical imaging and health informatics. It will provide clinicians with information that may be used to guide treatment or to match patients with targeted therapy options.
Just a few years ago, genetic studies were working with patient sample sizes in the low hundreds, and analyzing just parts of the DNA, says Brudno. Today, researchers are doing whole genome sequencing on thousands of patients, demonstrating that high performance computing has essentially become a basic research need.
“Through this collaboration, we will be significantly expanding our ability to handle data coming through existing genomics sequencing platforms by at least tenfold in the first phase of the project. The increased access to HPC will provide the Princess Margaret with the ability to process large and complex datasets in just hours as opposed to the current computational process time of weeks,” says Carl Virtanen, Bioinformatics Manager, Princess Margaret Cancer Centre, UHN. “By having a service such as HPC4Health available, we have guaranteed faster data processing which will allow expansion of clinical trials to include more patients overall.”
The HPC4Health infrastructure will be housed at and maintained by staff at the Scalar Decisions Datacentre at SickKids’ Peter Gilgan Centre for Research and Learning. This cloud-computing service offers a membership model that allows all partner hospitals to efficiently share resources including the temporary storage of sensitive data and dedicated staff with expertise in biomedical research. While HPC4Health will provide shared storage and resources, none of the organizations will have access to each other’s data, keeping personal health information protected and safe. HPC4Health is Compute Canada’s first health network node. Compute Canada coordinates and promotes the use of HPC in Canadian research to advance scientific knowledge and innovation.
HPC4Health will initially be available to hospitals in Toronto, but will be expanded to include healthcare organizations throughout Ontario and eventually across Canada. This health node will provide significant savings to all partners offering more access to high performance computing without the cost of building and operating their own IT infrastructure.
“HPC4Health is a model that will securely provide hospitals with data that allows physicians to make better decisions more quickly,” says Dan Sinai, chairman of Compute Ontario. “Centralizing resources is clearly needed to enable large-scale computing without compromising security. By providing a private cloud-based solution, HPC4Health will enable hospitals across Ontario to leverage expertise across institutions, improve diagnostics and lessen the burden on the health-care system.”
This project was supported by Compute Ontario, Compute Canada, the Canada Foundation for Innovation, The Princess Margaret Cancer Foundation, and SickKids Foundation.