SUNY at Albany
June 19-23, 2001
Statistical mechanics of proteins: perspective from analytical theory and simulations.
We will disscuss recent developments in protein theory that allow a unified statistical-mechanical view of protein folding and evolution. Specifically we will present a mean-field picture of protein evolution that, along with Monte-Carlo design in sequence space allows to rationalize most of the conservatism patterns in protein families and folds. Further, we will present novel Monte-Carlo simulations of protein folding in which all heavy atoms are represented as interacting hard spheres of various sizes corresponding to their van-der-Waals radii. This model includes all degrees of freedom relevant to folding - all sidechain and backbone torsions- and uses a Go potential. By recording many folding events over over a wide range of temperatures a possible folding mechanism for three different proteins - three-helix bundle, crambin and protein G is obtained. These results present a "proof-of-principle" for the possibility of a solution of protein folding problem at an all-atom level, provided that one has a realistic all-atom potential energy function that correctly favors the native state. In a separate development this algorithm helped to solve the ''side-chain packing problem'' by providing a rigorous estimate of the number of side-chain packing conformations that are compatible with a given backbone of a native protein.
Deaprtment of Chemistry and Chemical Biology,
12 Oxford, Cambridge, MA 02138,