Resisting Intuition: USU Scientists Use Computing to Predict Biology
Thursday, Dec. 06, 2012
USU scientists Greg Podgorski, left, and Nick Flann have developed varied computer simulations to predict biological processes, including angiogenesis in cancerous tumors. Their work was featured in a Nov. 2012 issue of the journal 'Nature.'
USU alum Art Mahoney '09, now a NSF Graduate Research Fellow at the University of Utah, helped his mentors Flann and Podgorski implement the computer model during his undergrad career.
Utah State University developmental biologist Greg Podgorski studies the processes by which cells develop into increasingly complex structures to form organisms. His colleague Nick Flann, associate professor in USU’s Department of Computer Science, delves into the science of computation and its role in biology. Together, the Aggie scientists explore the emerging fields of systems and executable biology. One of their recent collaborations, which involves modeling blood vessel formation in cancerous tumors, was featured in the Nov. 21, 2012, online edition of the journal Nature.
“Cancer is a complex disease that actually starts at a low genetic level but its outcome, as we all know, can be a major disruption at the multicellular level,” Flann says. “We’re trying to better understand what happens at the cellular level that alters what happens at the tissue and organ levels.”
Using an open source modeling environment called ‘CompuCell3D,’ Flann and Podgorski developed a model of a malignant tumor that allows them to explore angiogenesis, the physiological process involving the growth of new blood vessels from pre-existing vessels. Angiogenesis is of keen interest to cancer researchers because solid cancerous tumors rely on blood vessels to survive, grow and spread.
Solid tumors, which are not vascularized on their own, are aggressive recruiters of blood vessels, says Podgorski, associate professor in USU’s Department of Biology. “The tumors send signals to nearby blood vessels, which respond by growing new loops of vessels that reach out to the tumors and supply them with critical food and oxygen.”
At first glance, the solution to stopping the cancer appears to be cutting off its lifeline and many cancer treatment approaches – some in use, some under development – follow this logic. But cancer, and its underlying cellular processes, is not so simple.
“What we’re discovering from computer modeling is that cellular behavior isn’t always intuitive,” Podgorski says. “A cell signal at one level may trigger a myriad of reactions at multicellular levels that are far more complex and far-reaching than we could have anticipated and a non-intuitive approach may be warranted.”
Returning to the idea of cutting off a voracious tumor’s food supply: a more effective therapeutic approach, determined from the computer simulation, might actually be encouraging the malignancy’s appetite. The researchers reason the resultant increase in vascularization would amp the tumor’s ability to take in nutrition. Then, treatment would involve luring the hungry tumor to take a big, fatal bite of poison.
“This is just one scenario generated by our computer simulations,” Flann says. “As these ideas are generated we pass the information on to the scientific community, where other researchers explore treatment alternatives – such as drug delivery options to destroy growing tumors with as little harm to surrounding, healthy tissue as possible.”
He and Podgorski note a key feature of the computer model is its random generation of nearly endless cell growth scenarios.
“Each simulation unfolds differently,” Podgorski says. “The value of this is it would be impossible for humans to explore the virtually infinite, unpredictable possibilities of how angiogenesis could occur in response to signals from malignant cells.”
Interestingly, Podgorski and Flann mentored USU alum Art Mahoney ’09, now a National Science Foundation Graduate Research Fellow at the University of Utah who, as an undergraduate at Utah State, pioneered the development of the computer model.
“Art is an amazing scientist and we’re indebted to him for his work in implementing this project,” Flann says. “Without him, this work wouldn’t have been possible.”
What Mahoney and his faculty mentors recognized in developing the model is repetition is the key to discovery.
“We’re following Edison’s approach: trying a lot of different things and evaluating each,” Flann says. “Our advantage is we have access to supercomputing that allows us to do this very rapidly.”
To accomplish their simulations, the USU researchers secured parallel computing time – more than 4,000 computers running simultaneously – at the U.S. Air Force-funded Arctic Region Supercomputer Center in Fairbanks, Alaska and through NSF-supported TeraGrid resources provided by the Texas Advanced Computing Center in Austin.
The team’s research is supported by Seattle-based Institute for Systems Biology, as well as the Luxembourg Centre for Systems Biomedicine and the University of Luxembourg.
“Undergrad Researcher Receives National Recognition,” Utah State Today
Contact: Gregory Podgorski, 435-797-3712, email@example.com
Contact: Nicholas Flann, 435-797-2432, firstname.lastname@example.org
Writer: Mary-Ann Muffoletto, 435-797-3517, email@example.com