ITR: A Global Optimization Package for Protein Structure Prediction

University Of Colorado At Boulder, Boulder CO

Investigators

Abstract

Richard Byrd and Robert Schnabel of U. Colorado, and Teresa Head-Gordon and Silvia Crivelli of UC-Berkeley are supported by the Chemistry Divsion through the Information Technology Research Initiative to develop a global optimization package for protein structure prediction. Their approach combines optimization research from numerical computation, energy-modeling research from computational biophysics, and parallel computation research from computer science. Key features of this research include a novel representation of the aqueous solvent, incorporation of secondary structure prediction, and energy-smoothing models to increase accuracy and efficiency. The prediction of the three-dimensional shape, or native state, of a protein given its sequence of amino acids is one of the fundamental challenges of modern science. Protein structure knowledge is fundamental in understanding protein function, and important to many biomedical and biomaterial applications. This research will result in advances in large-scale and parallel numerical computation, and models and prediction methods for computational biophysics.