Entropy is a fundamental property of any system that quantifies the
+ structural flexibility of the system and together with energy governs
+ system stability. It is particularly important in biomolecular systems
+ because of their highly flexible and complex nature. Many properties
+ are a delicate balance of entropy versus energy, necessitating the
+ determination of entropy in order to understand stability. Moreover,
+ entropy provides a way to quantify the structural flexibility of a
+ system over all its degrees of freedom.
+
There are a range of existing methods to calculate entropy from + molecular dynamics simulations but they suffer from a number of + limitations: they may only work for particular kinds of system or + degrees of freedom, they may require additional calculations, they can + be difficult to interpret, or they do not scale well to large and + complex systems. Some methods only work for water, such as + Inhomogeneous Solvation Theory, for liquids such as 2-Phase + Thermodynamics, for only some degrees of freedom such as dihedral + binning, or for single molecules such as Quasiharmonic Analysis, + Normal Mode Analysis or non-parametric methods such as Minimal + Spanning Tree or K-Nearest-Neighbours.
+Given the widespread use of free-energy calculations and molecular + dynamics simulations, there is a large user-community for software to + calculate entropy and quantify full structural flexibility of + biomolecular systems. Multiscale Cell Correlation (MCC) provides a + novel solution to the problems encountered by other methods by + providing a single, scalable and general framework applicable to all + molecules in the system. It utilises a judicial synthesis of + mean-field cell theory and covariance matrices over a range of length + scales: First, Correlations are considered between groups of locally + connected atoms as in a mean-field cell, and longer-range correlations + are accounted for using a coarser representation of the groups, a + framework that is scaled to higher length scales. Second, At each + length scale, the potential energy surface is discretised into energy + wells for translational and rotational motion. These are represented + as an average energy well and an energy-well distribution, denoted as + vibrational and topographical, respectively. Finally, The + decomposition over molecules, length scales, type of motion and + energy-well size and distribution provides an exquisite level of + detail in explaining the entropies obtained.
+MCC has been applied by the group of RH to a wide range of systems, + namely liquids, aqueous and octanol solutions, host-guest complexes, + chemical reactions and large biomolecules such as proteins, DNA and + membrane bilayers in aqueous electrolytes.
+