Atomically Detailed Simulations of Heterogeneous Membranes
Thursday, April 16, 2020
3:30PM – 5PM
Biological Membranes are extremely heterogeneous materials. They are made of thousand types of phospholipid molecules and many inserts such as cholesterol molecules and transmembrane proteins. Membranes are also dynamics. Transient microdomain formations and the controversial concept of “rafts” are essential for biological functions such as signaling and transport. Atomically detailed simulations may shed light on these cooperative phenomena. However, Molecular Dynamics is too slow to observe significant assembly and dismantling events. Moreover, the acceptance probability of the usual Monte Carlo moves is also low in these dense systems. Therefore, the standard technologies for particle simulations fail to address membrane diversity.
I will present our new approach (MDAS) to simulate membranes, which is based on gradual Monte Carlo moves in which we accept or reject mutation trajectories. Provided that such trajectories can be designed, MDAS simulations are more efficient than conventional Molecular Dynamics by factors of thousands. I will illustrate the application of MDAS on a few model systems and discuss the phase diagram of a membrane that consists of DPPC and DLPC mixture.
Ron Elber is a W.A. “Tex” Moncrief Chair in Computational Life Sciences and Biology. He has been conducting Molecular Dynamics simulations of biological systems for as long as he remembers. He has been influential in the field and introduced several widely recognized and used algorithms for particle simulations. Among those are algorithms to compute reaction paths based on a whole curve optimization, and the use of phase space partitions to conduct simulations of rare events. He is working on proteins, RNA molecules, and membranes, but so far did not touch sugars. The talk will describe a recent algorithmic addition to the simulations of heterogeneous material and its applications.
Please join this Zoom seminar online with the "Audio Only" function (no video).