Protein structure/hit-and-run sampling

Justin Melvin (Merrick Furst) - Protein structure/hit-and-run sampling
One of the major unsolved problems in biology is the protein folding problem: given a protein as a sequence of amino acids, find the biologically relevant "fold": the 3-d structure that the protein contorts to in the cell. The structure of a protein is vital to its function; moreover, given the structure, biologist can often guess the function of a protein. This is especially relevant given the huge number of proteins whose amino acid sequence is know (for example, from the Human Genome Project), but whose function is not. Among the problems facing current approaches is the difficulty in sampling from the configuration space of all possible folds. In this research, we apply recent advances in efficient sampling from a convex set to the protein folding problem.