University of Delaware campus
Published:
9/11/2019

FREDERICK, Md. -- High-performance computing and other data science technologies are enabling critical inroads in cancer research. However, bringing together the necessary computing and cancer biology expertise to effectively leverage these technologies is a challenge—a challenge that a new partnership between the Frederick National Laboratory for Cancer Research and the University of Delaware is working to overcome. 

The Frederick National Laboratory (FNL), operated by Leidos Biomedical Research, Inc. for the National Cancer Institute, and the University of Delaware recently signed a memorandum of understanding to explore avenues for collaboration around using computational and data science to tackle cancer, AIDS, and emerging health challenges.

Eric Stahlberg, Ph.D., director of FNL’s Biomedical Informatics and Data Science (BIDS) Directorate, spearheaded the collaboration along with Sunita Chandrasekaran, Ph.D., assistant professor of computer and information sciences at the University of Delaware.

The two parties identified several areas for potential collaborative research, including high-performance computing to accelerate genome mapping and assembly, data science, and the use of field-programmable gate arrays (FPGAs) for artificial intelligence, imaging analysis, and additional areas of study. FPGAs are integrated, customizable circuits, and they hold significant potential for enabling applications to run faster and at lower power requirements.

Stahlberg explained that the University of Delaware is making contributions at the forefront of high-performance computing in key areas of interest, while BIDS has expertise in applying these technologies to cancer research. 

“It’s a natural fit,” he said, “where the community will benefit as the limits of high-performance computing applied to cancer research continue to be pushed.”

Chandrasekaran said she was “very much looking forward to developing collaborative research and training activities with FNL to drive innovations and development in the field of biomedical research.”

The FNL is engaged in several other collaborations utilizing high-performance computing, including the Cancer Distributed Learning Environment (CANDLE) and Accelerating Therapeutics for Opportunities in Medicine (ATOM) public–private consortium. 

Through CANDLE, the FNL, in collaboration with four Department of Energy national laboratories led by Argonne National Laboratory is developing an open-source software that can help predict which drugs would be most effective against specific cancers. Meanwhile, ATOM seeks to leverage high-performance computing, shared biological data, and emerging biotechnologies to accelerate the discovery of effective cancer therapies.

The new partnership with the University of Delaware helps to broaden the community and will eventually tie into these or other established collaborations. 

“There is definitely a need for collaboration when computing, big data, and science challenges are brought together,” Stahlberg said. “It is difficult to have expertise in all areas needed for successful collaborations. A much more effective approach is to build the community across the disciplines, so that becomes a priority. We work to bring together those with interest in the science challenges posed in cancer research and the opportunities found in applying new computing technologies to make progress.”

Creating cross-organizational teaching and training opportunities is another major element of the new partnership. The FNL and University of Delaware will work to provide training opportunities for students to apply their classroom experience with high-performance computing to real-world biomedical challenges. Two students from the University of Delaware have already worked with BIDS on CANDLE challenges, and the new agreement will open the door for expanded opportunities and interactions. 

By Victoria Brun, Partnership Development Office

Image: University of Delaware campus, public domain image, Wikimedia Commons

Tagged: