Toward accurate coarse-graining approaches for protein and membrane simulations

Abstract

In the last few years, several multi-scale approaches aimed at overcoming time- and size-bottlenecks in atomistic simulations have been proposed. Amongst them, coarse-graining methods appear the most promising ones in terms of compromise between computational cost and molecular accuracy. Because of their success, a large family of coarse-grained models addressing different soft matter systems are available in the literature and they are nowadays commonly used to study complex phenomena such as molecular aggregation, protein folding, or DNA assembly. In this chapter, we restrict our analysis on current methodologies used in protein and membrane simulations using multi-scale and coarse-grained potentials. We introduce fundamental concepts in coarse-grained modelling, and provide examples of some of the most widely used forms of Hamiltonians. Among others, we discuss elastic-network models as well as the MARTINI and UNRES force fields for proteins. We also briefly discuss parameterisation protocols for coarse-grained potentials, including iterative Boltzmann inversion and force matching. In the second part of the chapter, we review current approaches to biological membrane simulations, comparing atomistic, coarse-grained and continuum models. Particular attention is given to the MARTINI and SDK potentials. Alongside with presenting the available coarse-grained models, we discuss their current limitations and the challenges to improve their reliability.