Coarse graining: a tool for large-scale simulations or more?

Reduction of the representation of macromolecular systems such as, e. g. biopolymers from an all-atom to a more crude one, which is termed coarse-graining, is usually regarded as a means to extend the time-and size-scale of simulations. The respective effective energy functions can be defined as potentials of mean force (PMF) of the respective systems, in which the degrees of freedom not present in the reduced model are integrated out. In this comment it is demonstrated, with the example of the UNRES (UNited RESidue) model of proteins, that analysis of the components of the PMF, after its expansion into cluster-cumulant functions, enables us to use coarse graining to understand the origin of macromolecular structure. It is also suggested that mean-field interactions between the multipole moments of polar units in the context of local interactions are the key factor that determines basic regular structural patterns of biopolymers.