There is considerable interest in the lateral organization of biological membranes. While membrane proteins obviously influence on the localized properties of membranes, the lipids themselves also can self-organize within the bilayer. Computational models can complement the growing body of experimental studies of the properties of heterogeneous lipid membranes. In this talk I will describe work from my lab and the lab of Eric Jakobsson at the University of Illinois which has focused on using computational modeling to understand the molecular interactions between lipids that underlie phase separations and domain, or `raft'', formation and stability in mixed lipid bilayers. I will begin with a brief review of earlier atomistic simulations of mixed lipid membranes. I will then describe how we have used data from simulations as input to a dynamical Mean Field model for mixed lipid bilayers that allows the modeling to be extended to biological length and time scales. This approach has been applied to DPPC-Cholesterol mixtures, and we are now extending the calculations to DOPC-Sphingomyelin (SM) and DOPC-SM- Cholesterol mixtures. I will conclude with a description of the most recent work from my lab, including a new Monte Carlo approach to the atomistic simulation of mixed lipid bilayers.
# Self-consistent Mean Field Model Based on Molecular Dynamics: Application to Lipid- cholesterol Bilayers.'' , with G. Khelashvili and S. Pandit, J. Chem. Phys. 123, 034910 (2005)
# Sphingomyelin-Cholesterol domains in phospholipid membranes with S. Pandit, R. Jay Mashl, S. W. Chiu, S. Vasudevan, and E. Jakobsson, Biophys J, J 87, 1092-1100 (2004)
# Simulation of early stages of domain nano-domain formation in mixed lipid bilayers of sphingomyelin, cholesterol, and dioleylphosphatidylcholine”, with S. Pandit and E. Jakobsson, Biophys J. 87, 3312 - 3322 (2004)
# Modeling the lipid component of membranes , Curr. Op. Struct. Biol. 12. 495-503 (2002)