How optimized GNM force constants vary with inter-residue distance and contact order.

Their conceptual simplicity, exact results and ease-of-use make elastic network models attractive for investigating protein structural dynamics. Using little more than structural information, these models are abel to predict large-scale motions of proteins, and their success underscores the role of native contact topology as the primary determinant of global equilibrium dynamics. With the accumulation of structural data, including in particular ensembles of structures resolved for the same protein, we are in a position to systematically examine the experimentally detected variations in structure and unravel for the first time the more subtle effects - beyond native contact topology - that influence the experimentally observed dynamics. We focus in particular on cross-correlations between residue fluctuations, which have not been rigorously examined in previous studies of equilibrium dynamics, and we assess the role of various sequence- and structure-dependent characteristics in defining the collective changes in structure. Notably, geometry-based factors, such as contact order and secondary structure, are distinguished as properties affecting the dynamics, whereas residue-specific effects are less influential.

By applying the principle of entropy maximization to an elastic network model of predetermined topology, we were able to calculate the values of the force constants that account for the observed cross-correlations between residues. Our research demonstrates how information gleaned from experimental data can be rigorously incorporated into coarse-grained models, which is important because much of the conformational space of equilibrated proteins remains beyond the reach of all-atom models, but can be explored through simpler, coarse-grained models with the proper parameterization.

The figure shows how optimized GNM force constants vary with inter-residue distance and contact order.

Related publication: Lezon TR & Bahar I. (2010). Using entropy maximization to understand the determinants of protein structural dynamics beyond the native contact topology. PLoS Comp Biol 6: e1000816.

Back