Gaming technology brings biology models to life

Nancy D. Lamontagne

The graphics processing units (GPUs) behind today’s quick and complex gaming systems are being harnessed by researchers at Michigan Technological University in Houghton to build simulations of complex biological systems.

Researchers used an agent-based model to simulate the human response to tuberculosis. First a nodule called a granuloma forms containing macrophages. At the core of the nodule are macrophages infected by tuberculosis.

They are using agent-based modeling, which predicts outcomes by simulating the behaviors of complex systems. In biology, this type of approach can be useful because expensive and time-consuming experiments can be simulated before they are done in real life. It can keep scientists from reaching dead ends and, in some cases, can enable even more complex tests than can be performed in real life. For example, instead of knocking out one gene in a mouse, a simulation could examine the consequences of knocking out multiple genes in humans.

This type of modeling has been around for some time but had been held back by computing power: The only way to run the models quickly was by using multimillion-dollar supercomputers. However, because graphics processing units are adept at crunching data quickly, the researchers, led by Dr. Roshan D’Souza, used them to run a simulation of the human response to tuberculosis developed by Dr. Denise Kirschner of the University of Michigan in Ann Arbor. The resulting 3-D simulation of T cells and macrophages is produced from millions of real-time calculations. D’Souza said that their method is several orders of magnitude faster than current modeling toolkits.

Next, T cells (shown in purple) are recruited.

The researchers plan to examine even more complex systems using this technique, including modeling how tuberculosis could spread from the lung to a patient’s lymphatic system, blood and vital organs.

Finally, T cells enter the granuloma and kill the infected macrophages. Studying virtual biological systems such as this will not replace the laboratory, but it could help scientists focus their studies on the right things.