First principles based molecular dynamics simulations can nowadays give insights into atomistic mechanisms of many complex phenomena but in some cases, the use of high-level electronic structure methods imposes severe limits on the tractable system size and the accessible time scale. These limitations are especially cumbersome in the simulation of biological systems that often contain 10’000 to 100‘000 of atoms and involve typical time scales that can exceed even milliseconds. In addition, functional and non-functional conformations of biological systems can lie in a very narrow energy range setting stringent requirements for the accuracy of a computational method. In this talk, some typical examples of biological problems will be discussed that illustrate the need for high-accuracy methods. Different approaches that can potentially address this combined challenge of extended size, long time scales and high accuracy such as the on-the-fly adaption of classical force fields [1,2] and semiempirical methods [3] based on data from high-level quantum chemical methods will be presented together with efficient QM/QM and QM/MM methods based on multiple time step integrations. [1] P. Maurer, A. Laio and U. Rothlisberger, Automated Parameterization of Biomolecular Force Fields from QM/MM Simulations through Force-Matching, J. Chem. Theor. Comp. 3, 628-639 (2007) [2] M. Doemer, P. Maurer, P. Campomanes, I. Tavernelli, and U. Rothlisberger, A generalized QM/MM Force Matching Approach applied to the 11-cis Protonated Schiff Base Chromophore of Rhodopsin, J. Chem. Theor. Comp. 10, 412-422 (2014) [3] M. Doemer, E. Liberatore, I. Tavernelli, and U. Rothlisberger, In situ parameterization of SCC-DFTB repulsive potentials by iterative Boltzmann inversion, Molecular Physics 111, 2595-2607 (2013)

Multiple time step schemes for first-principles based multiscale simulations

251st National Meeting (2016)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.