Meet a RC Researcher- Nicole Speer

Nicole Speer is the Director of Operations for the Intermountain Neuroimaging Consortium, in which neuroscientists from across the Rocky Mountain region study psychological processes such as addiction, pain, emotion, attention, sleep, learning and memory. Speer earned her Ph.D. in psychology and held a postdoc position at the University of Colorado, Boulder, from 2005 to 2007, during which she completed her own brain imaging research. She returned to the university four years ago to manage the MRI scanner for the brain imaging research center. In her current position, Speer says, “I make sure everyone has access to the resources they need and that people are able to finish their studies on time, and I deal with any problems that arise.”

Research in Speer’s facility begins with brain imaging from MRI scans. Subjects lay in the MRI scanner with a coil around their head that amplifies the signals their brains produce. The brain images collected in scans can be structural, showing the locations of the hippocampus, motor cortex and cerebellum, as well as the amounts of grey and white matter. “You might look at those white matter tracks in the brain and see if they’re different in people who go on to develop psychosis than in those who don’t,” Speer says. The bulk of the images that her researchers collect examine activation in the brain by measuring blood flow between 3-cubic-milimeter voxels. “You’re generally putting those voxels together, because if just one little tiny voxel in the brain shows a change, that doesn’t necessarily mean much,” Speer explains. “If that voxel and 20 of its closes neighbors show the same pattern, you can be a little more confident that you’ve found something.” Blood flow data is mapped onto a high-resolution image of the structure of the brain to show researchers the locations of these changes in blood flow. Many studies examine brain activity as participants read or listen to music. “Other labs are interested in things you can’t see or hear,” Speer says. “We have a lab that studies pain and how pain is perceived, so they put little devices on people that administer pain.” Watching the brain respond to pain has enabled researchers to develop an algorithm for the amount of pain a person is experiencing, creating a more objective measure for medical professionals than asking their patients to describe the amount of pain they feel.

As researchers collect brain images, they process individual participants’ data. Once their studies are finished, they begin combining those processed data into larger analyses so that they can look for consistent patterns in brain activation. Certain researchers have begun incorporating genetic data into their analyses, which adds another level of complexity: rather than simply asking whether a certain part of the brain is active while a subject is reading, researchers are trying to determine whether the level of activity in a part of the brain differs depending on whether the subject possesses a certain gene. Studies range in size from 20 participants up to 600, and the complexity of brain image data requires computational infrastructure for storage and analysis. Researchers store their data in the PetaLibrary, from which they can access it with processing nodes in the Blanca cluster for analysis. Research Computing manages both Blanca and the PetaLibrary.

Using shared Research Computing resources, rather than using their own computing systems, is enormously beneficial to researchers. “Especially as people’s data sets are getting more complicated, with more variables and more people who are being followed for longer periods of time, it doesn’t make sense for each lab to get these huge computing systems to store and analyze their data,” Speer says. “People are coming to embrace Research Computing as the place to get their research done, and I think that’s really a good thing.” Sharing resources is also beneficial because it encourages collaboration between labs. When one lab uses shared resources to develop an especially effective way to process data, that processing pipeline becomes easily accessible to other labs. As a facility manager, Speer also appreciates Research Computing’s role in maintaining resources, which enables her to spend her time supporting research rather than troubleshooting. “It’s a relief to have that in someone else’s hands. There’s one point of contact for any problem that comes up: Research Computing handles it all, and that’s really nice,” Speer explains. “I absolutely love this model. The more we can shift joint things to other groups in the university, the more benefits we see.”

Speer’s researchers started with three nodes on Blanca, and have been able to add two more nodes in the past four years. “That’s the other nice thing about having Research Computing take this on: they can expand much more easily than we could,” Speer says. If she and her team were maintaining a server on their own and wanted to expand, they might have to obtain an entirely new system. Research Computing has the capacity to simply connect extra nodes. “It doesn’t take a lot of effort from us to expand, and that’s really nice, especially as these studies keep getting bigger and more people are doing imaging studies.” The nature of modern brain imaging requires that researchers utilize computing resources. “If you’re doing scanning now, you learn about Research Computing. They go hand-in-hand,” Speer says.

The researchers Speer oversees are using these resources for a variety of exciting projects. One project, which was recently submitted for publication, examined white matter tracks in the brain. White matter tracks, which connect different regions of the brain, develop as children age. They grow especially quickly when puberty begins, replacing the grey matter that connects individual neurons. The researchers examined the impact of the age of first marijuana use on the growth of white matter tracks and found that the earlier a subject began using marijuana, the more disrupted these white matter tracks were. Although data for this study was collected at a different site, analysis was run at the University of Colorado because of the computing resources available on campus. “That’s a benefit that comes out of our ability to run these large analyses on really big data sets,” Speer points out.


Another study at Speer’s facility is taking advantage of the twin study at the Institute of Behavioral Genetics. The 600 twins in this study have been followed for over 20 years, and researchers are currently collecting images of their brains. “They can see a lot of really interesting information from comparing, say, how a person performed on a memory task when they were six years old with what their brain looks like now or how their brain responds to a particular task when they’re 25,” Speer explains. The study is examining executive function, participants’ ability to control their behavior and attention. Other researchers are running intervention-type studies. These studies begin by scanning participants’ brains to collect information on structure and the way brains respond to stimuli. They then introduce an intervention, such as regular exercise, mindfulness or nutritional supplements and examine the intervention’s impact on the brain’s structure and response to stimuli. Additional studies examine childhood brain development and the way it is influenced by a variety of factors, such as poverty. Speer says, “All of our research projects are very interesting and exciting, thanks to the resources we have available.”