Folding of villin protein simulated on GPUs
This movie shows a possible trajectory, or path, for how the villin protein folds. A single trajectory can be generated in a few days on a graphics processing unit (GPU) or a year on a single CPU. However, protein folding is highly stochastic, or random. To actually understand the folding mechanism, thousands of trajectories have to be simulated. The underlying data to produce this typical trajectory required about 20 days on 1000 GPUs. It was accomplished via Folding@Home, a distributed computing project by Stanford faculty Vijay Pande where people from throughout the world download and run software to band together to make one of the largest supercomputers in the world. Access trajectory files for villin from simtk.org Learn more about Simbios research on physics-based simulations of biological structures at simbios.stanford.edu.
