Mechanics and Computation / View source

for "Longer Timesteps for Molecular Dynamics of Proteins using Normal Modes"

Longer Timesteps for Molecular Dynamics of Proteins using Normal Modes

Molecular Dynamics (MD) are an important tool for the study of protein folding, conformational change, and other applications. 
MD is fundamentally limited because of the short timestep that needs to be used, in femtoseconds, to account for fast vibrational motions. I describe progress in using normal mode decomposition of MD to more quickly sample and study dynamics of proteins. The method periodically computes a fast approximation of the slow modes, propagates in the slow mode space, and then uses a Monte Carlo method to sample the space of the fast modes. Numerical results and performance results from implementations in GPUs will be presented. 

Jesus A. Izaguirre is Asssociate Professor of Computer Science and Engineering at the University of Notre Dame. He has over 15 years of experience in methods for accelerating and improving MD simulations. He also works on applications of these technologies in protein folding, rational drug design and computational biology. His research is funded by the NIH and NSF. During 2014 he is a Visiting Scholar in the Dept. of Chemistry at Stanford University, collaborating with Prof. Vijay Pande, and also collaborating with Prof. Eric Darve at ICME. He is also founding CEO at Oak Financial Software, a startup in Silicon Valley. He can be reached at izaguirr@stanford.edu

Return to "Longer Timesteps for Molecular Dynamics of Proteins using Normal Modes".

Retrieved from "http://mc.stanford.edu/%22Longer_Timesteps_for_Molecular_Dynamics_of_Proteins_using_Normal_Modes%22"