Monday July 25/1:30/Grande Ballroom

Invited Presentation 3

Chair: Bo E.H Saxberg, Eli Lilly & Co.

Molecular Dynamics and Supercoiled DNA

The importance of biomolecular modeling and simulations is steadily increasing with progress in experimental techniques, algorithm development, and new computing technologies. However, exceptional progress has been limited by two major problems: uncertainties in force fields and inadequacy of configurational sampling for large systems. The sampling problem is being addressed through faster computers and powerful simulation techniques including global optimization, Monte Carlo, molecular dynamics, and Brownian dynamics. In theory, molecular dynamics simulations -- in which atomic motion is propagated through numerical integration of the classical equations of motion -- can bridge the spatial and temporal resolution and thus capture molecular motion over a wide range of thermally accessible states. In practice, the numerical timestep problem has limited most applications to integration with very small timesteps compared to the scale of motion of interest (e.g., protein folding).

The speaker will present a technique that combines implicit-integration with normal-mode methods. The high-frequency and rapidly-varying components of the motion are resolved in the normal-mode phase, by solving the linearized Langevin equation; the slowly-varying parts are resolved by a large timestep implicit integration step. Applications of the algorithm to supercoiled DNA -- a higher-order form of the DNA in which bending about the global helical axis occurs -- reveal interesting large-scale features (e.g., bending and twisting) involved in the folding of DNA and its interactions with regulating proteins.

Tamar Schlick, Chemistry Department, Courant Institute of Mathematical Sciences, New York University

Tamar Schlick received her B.S. degree in Mathematics from Wayne State University in 1982 and her M.S. (1985) and Ph.D. (1987) degrees in Applied Mathematics from the Courant Institute of Mathematical Sciences, New York University. Following a two-year NSF Postdoctoral Fellowship at NYU and the Weizmann Insitute, she joined the faculty of Chemistry and Mathematics at NYU, where she is currently an Associate Professor.

Professor Schlick is an Alfred P. Sloan Research Fellow, an NYU Distinguished Recent Alumna, a NSF Presidential Young Investigator, a Whitaker Biomedical Engineering Fellow, and a Searle Scholar. Among the many awards she has received are the K.O. Friedrichs Prize for Outstanding Dissertation in Mathematics, Courant Institute, and the J. Krakauer Prize for Outstanding Dissertation in the Sciences, NYU, both in 1988.

Her research interests include algorithm development for biomolecular simulations and their application to DNA and protein structure, in particular the study of supercoiled DNA folding and knotting. Her other areas of interest are nonlinear unconstrained optimization, especially truncated Newton algorithms, stiff differential equations, conjugate gradient methods, application tailored preconditioners, modified Cholesky factorizations for indefinite preconditioners, sparse linear systems, and sparse spectral decompositions.