Video games assist biology research
What is a graphic processing unit to you? An avid gamer may know that it is the equipment behind realistic graphics on a Playstation or an Xbox. However, that is not all a graphic processing unit (GPU) can be used for. In fact, the university’s own Samuel Cho, assistant professor of physics and computer science, is using GPUs to aid him in research.
Although the technology has been successfully employed already in the fields of bioinformatics, finance and physics, Cho was eager to try it with his own algorithms.
It was with GPU technology that China displaced the United States for the number one spot on the Top 500 Supercomputer list in November 2010, which is compiled by various academic institutions.
- Photo courtesy of WF News Center
With the help of Anqi Zou, a senior mathematical business and computational science major at the university, Cho simulated models of human molecular and enzymatic processes.
Cho encourages Zou to participate in research, saying that when students are involved, “our research performance becomes better and our students become more engaged.”
“The most important part of solving a problem is to solve the problem as completely and simply as possible,” Cho said.
Although he is not a professor of biology, Cho is still interested in the topic. Similarly, Cho has proven to be a mentor for Zou in research outside of her field of study. To exemplify this interdisciplinary collaboration, Cho points out that “about 30 percent of all physics majors also major in a different discipline. Even our own Dean of the College, Jacque Fetrow is a professor of physics and computer science.”
Last spring, the university was declared a CUDA (Compute Unified Device Architecture) Teaching Center by NVIDIA Corporation, the mother company that donated the GPU technology to the university.
CUDA is the parallel computing architecture through which GPU technology is employed. Cho is the CUDA coordinator and is teaching a new course in the computer science department called “GPU Programming.”
In Cho’s most recent published study out of the University of Maryland, Cho has created a model of the human telomerase enzyme, a crucial RNA molecule in cancerous cells. His team created a working model, simulating telomerase folding and unfolding.
Human telomerase exists only in cancer cells and is responsible for the uncontrollable division that causes tumors. If the human telomerase molecule is understood, its function can be limited. Thus, limiting the activity of human telomerase has the potential to aid in cures for cancer and tumor therapies.
Additionally, Cho and his assistants are looking at bacterial ribosome, a much larger molecule than human telomerase. With the use of a standard computer, this type of simulation could take over 40 years. However, with GPUs they will see results in a matter of months.
Ultimately, Cho and his team hope to better understand the functions of the ribosome in order to develop specific antibiotics to target it.
Cho thanks gamers for buying the systems because without them “the prices wouldn’t have dropped.”
A few years ago, GPUs were far too expensive to be used for science, but due to supply and demand the technology is now far more affordable and can be used to tackle problems much larger and more complicated than ever before.
In the realms of physics, chemistry, biology, and economics, GPUs can prove to be a stepping-stone toward mastery and understanding.