Title: Multiscale Modeling of Complex Molecular Systems

Abstract

This presentation concerns the in-silico study of complex molecular systems at different levels of description and is divided into two parts. The first one focuses on polymer and polymer-based systems simulated by a mixed particle and field-based Brownian dynamics/kinetic Monte Carlo simulation approach that is fully implemented by the Engine for Mesoscopic Simulations for Polymer Networks (EMSIPON) written in C++. It is pointed out that EMSIPON has been entirely developed in the Computational Materials Science and Engineering Group of Professor Theodorou at the National Technical University of Athens. From the thermodynamic point of view, the systems are fully described by a Helmholtz energy consisting of various contributions: entropic springs, slip springs, nonbonded interactions and interactions between the polymer coarse-grained sites and solid inclusions (e.g. nanoparticles). Parameters invoked by the mesoscopic model are derived either from experimental data or from atomistic simulations establishing thus a bottom-up framework. The presented in-silico approach is employed for the calculation of structural, dynamical, thermodynamic and viscoelastic properties.

The second part of the presentation is devoted to biased and unbiased atomistic molecular dynamics simulations of central nervous system drugs in cholesterol-free and cholesterol-containing hydrated lipid bilayers, as well as in aqueous solutions containing β- cyclodextrin which is a cyclic oligosaccharide employed as a carrier in drug delivery systems. From the biased simulations, significant thermodynamic properties are computed, such as the free energy difference of partitioning into the lipid phase, the free energy barrier for hopping events between the two leaflets of the bilayers and the most-preferable position of the drug along the selected reaction coordinate which is the axis perpendicular to the two leaflets of the lipid membrane. The binding free energy of the noncovalent complexes formed between the drugs and β-cyclodextrin is estimated by applying the linear interaction energy method. In addition, several other properties are calculated by the unbiased simulations.