Correlation analysis helps to determine connections between the movements of residues. In many cases correlated movements of two residues result from trivial relationships (such as sequence connections), but in others signals biologically important connections.
The analysis in this page is a simple view of the correlation matrix raw data. A square matrix is presented in the page, where each row and column represents a residue. Each position shows the correlation value for that pair of residues, and the cell is coloured with a white to green gradient for positive correlation and white to red for negative ones.
The user can selected to apply a basic filter to remove large but irrelevant correlations from the analysis. This is done by applying a Z-score filter (D’Abramo et al., to be published), which signals when a residue-residue correlation is higher (in absolute value) than that expected considering the inter-residue distance. The background model used to compute the Z-score was obtained from the analysis of atomistic molecular dynamics trajectories of all protein metafolds (Rueda et al. 2007b). The filter eliminates irrelevant correlations due for example to sequential connections. In the filtered plots, small, and large but not significant correlations are coloured in white
Correlations are indicated as a 2D NxN plot. The exact value of the correlation can be recovered by placing the mouse over the appropriate position of the graph. When more than one chain is involved in the study, chain boundaries are indicated with black lines in the graph.