"The traditional computational modeling of protein structure, dynamics, and interactions remains difficult for many protein systems. It is mostly due to the size of protein conformational spaces and required simulation time scales that are still too large to be studied in atomistic detail. Lowering the level of protein representation from all-atom to coarse-grained opens up new possibilities for studying protein systems. In this review we provide an overview of coarse-grained models focusing on their design, including choices of representation, models of energy functions, sampling of conformational space, and applications in the modeling of protein structure, dynamics, and interactions. A more detailed description is given for applications of coarse-grained models suitable for efficient combinations with all-atom simulations in multiscale modeling strategies."
abstract of our
The review paper on Coarse-Grained Protein Models and Their Applications [pdf]
Practical applications of classical atom-level molecular modeling are still limited by its algorithmic efficiency and the available computing power. Even using a special-purpose supercomputer dedicated to atomistic molecular dynamics (MD) simulations, it is possible to simulate folding processes of only small, relatively fast folding proteins or their dimerization processes. Similar limitations apply to molecular docking, studies of dynamics of biomacromolecular systems, and other related tasks. This is a major reason why development and practical applications of coarse-grained protein modeling methods is needed. Application ranges for molecular modeling at different resolutions - quantum, all-atom, coarse-grained, and mesoscale - are presented below. The figure shows approximate ranges of time scales and system sizes (lengths).
The figure from Chem Rev, 2016, doi:10.1021/acs.chemrev.6b00163
A broad spectrum of coarse-grained protein chain representations have been developed so far. In all cases, the main purpose was to reduce the number of degrees of freedom treated in an explicit way. For site this reason, pseudoatoms replace amino-acid fragments (an equivalent term used in literature for “pseudo-atom” is “united atom”). Below, all-atom representation of a tripeptide and the corresponding various coarse-grained models are presented: Rosetta centroid mode (CEN) representation, CABS, UNRES, SICHO, and Levitt and Warshel model. United side chain atoms are colored in orange. Pseudobonds of fluctuating length are shown as springs and lattice models are shown on the underlying lattice slide.
The figure from Chem Rev, 2016, doi:10.1021/acs.chemrev.6b00163
Designing force fields for coarse-grained models is to some extent directed by the chosen level of resolution and the expected ranges of applicability. Typically, in comparison to its all-atom counterpart, the coarse-grained force field smoothens out the energy landscape, and thereby helps to avoid local energy minima “traps,” The figure below illustrates the effect of the smoothening of the energy landscape in a coarse-grained model as compared to an all-atom model. The flattening enables efficient exploration of the energy landscape in search for the global minima, while avoiding traps in the local minima.
The figure from Chem Rev, 2016, doi: 10.1021/acs.chemrev.6b00163
Short slide presentation of the review paper: