A Probabilistic Framework for the Characterization
of the Protein Native State
Wednesday, March 16, 4:00 PM
Packard Lab room 466
Reception prior to talk in Packard Lobby
Abstract: Many search and optimization problems exhibit complex high-dimensional non-linear spaces. Protein systems are ubiquitous biological molecules characterized by such spaces. A fundamental issue in our understanding of biology and treatment of disease concerns elucidating the structures and motions that proteins employ for their biological
function in the living cell. Doing so in silico involves exploring a high-dimensional conformational space and a continuous rugged energy surface associated with this space in search of the protein native/biologically-active state. This talk will present a robotics-inspired probabilistic framework to enhance the sampling of biologically-active conformations. Novel strategies are introduced to handle the high-dimensionality of the search space. Energy- and geometry-based discretization layers are employed to gather information about what is already explored in order to further guide the search towards relevant regions of the conformational space. Extensive applications suggest the proposed efforts greatly enhance the sampling of the protein conformational space and efficiently recover functionally-relevant conformations. Interesting insight is obtained on how to tackle the dimensionality challenge both in protein chains and articulated mechanisms.
Bio: Amarda Shehu is an Assistant Professor in the Department of Computer
Science at George Mason University. She holds affiliated appointments
in the Department of Bioinformatics and Computational Biology and the
Department of Bioengineering at George Mason University. She received
her Ph.D. in Computer Science from Rice University in Houston, TX in
2008, where she was an NIH fellow of the Nanobiology Training Program
of the Gulf Coast Consortia. Her research interests encompass
computational structural biology, biophysics, and bioinformatics.