Abstract

Coarse-grained approaches, in which groups of atoms are represented by single interaction sites, are very important in biological and materials sciences because they enable us to cover the size- and time-scales by several orders of magnitude larger than those available all-atom simulations, while largely keeping the details of the systems studied. The coarse-grained approaches differ by the scheme of reduction and by the origin and parameterization of the respective force fields. Both statistical (database-derived) and physics-based potentials are used, the physics-based potentials enabling us to bridge the coarse-grained level with the all-atom level, which is necessary when carrying out the simulations at multiple resolutions (multiscale simulations). The physics-based potentials originate from the potential of mean force (PMF) of a system under study, in which the degrees of freedom that are not considered in the model are averaged out. For tractability and transferability the PMF has to be expressed as a sum of contributions that constitute the effective energy terms. These terms are often assigned analytical expressions imported from all-atom force fields or engineered to reproduce certain structural patterns (e.g., the secondary structures of proteins or nucleic acids). Tabulated (model-free) potentials are also applied. Approaches also exist in which the effective energy terms are derived systematically by splitting the potential of mean force into transferable terms, e.g., by expressing the PMF by the Kubo cluster-cumulant functions. Two approaches, or a combination thereof, are applied in the parameterization of the coarse-grained force fields: the bottom-up one, in which the potentials of mean force are determined from atomistically-detailed calculations and then used to parameterize the respective expressions, and the top-down approach, in which the force field is tuned to fit the experimental data. In this chapter, the theory and parameterization of the physics-based coarse-grained force fields, along with the corresponding methods of conformational search are reviewed. Examples of physics-based coarse-grained force fields applied to study biomolecules and their assemblies and nanosystems are discussed.

Citations

Authors (8)