Dynamics of Tsr dimer.

<p><b>A</b> The ensemble of structures from a single simulation, fitted on the hinge region (marked). Individual structures are coloured by timestep; there is no clear pattern in kinking over time. To find the hinge point of this bending motion we calculated the mean-square-deviation of CA positions from two lines of best fit of different bisections of the protein around the HAMP/coiled-coil domain interface. <b>B</b> First mode of a GNM analysis of the dimer model, indicating a bending motion around the same hinge observed in MD. The dominant eigenvector (consisting of a scalar value for each residue) is plotted as transparent sphere overlaid on the cartoon representation of the structure, with red/blue indicating positive/negative scalar values. The hinge region is identified from the area with small eigenvector values, following <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1002685#pcbi.1002685-Kundu1" target="_blank">[58] </a><b>C</b> Histogram of bending angles from a single Tsr simulation. The modal angle is roughly 30°. <b>D</b> Bending analysis of the helix. 2 lines of best fit are generated from backbone particles above or below the identified residue, and the mean square deviation of fits of these 2 axes is calculated and plotted. The minimum represents the location of the bending hinge (i.e. the point where the backbone positions best fit a line).