Skip to main content
Figure 2 | Genome Biology

Figure 2

From: How biologically relevant are interaction-based modules in protein networks?

Figure 2

Modular organization, mean similarity dendrogram and phylogenetic profile. Modular organization, mean similarity dendrogram and phylogenetic profile of (a-c) cellular rescue, and (d-f) cellular environment functional networks. (a-d) Modular organization extracted with the network clustering algorithm. Protein interactions are plotted in brown. Modules are highlighted in white. Proteins within each module have been reorganized to show those with the greatest intra-modular connectivity - the core proteins - in the center of the module. (b,e) Mean similarity dendrograms. Branches for each corresponding module in (a) and (d) are joined at a node plotted at . Branches terminate at the mean similarity of each module, W m , giving branch lengths of W m - in similarity units. Dendrograms related to full modules are in black and those corresponding to the core components are in red. Those branches statistically significant (P < 0.05) end in a circle. (c,f) Continuous phylogenetic profiles color-coded from dark blue (maximal homology) to brown (no homology). Columns show the presence or absence of network nodes in a given organism and rows show the presence or absence of a given node in all the organism set. Species are arranged in taxonomic groups separated by white dashed vertical lines: Bacteria (left), Archaea (center), and Eukarya (right) (see Additional data file 1). The horizontal white dashed lines represent the localization of modules. A quick look at these figures provides evidence that proteins that are part of the same module exhibit a loosely correlated degree of conservation, as should be the case if modules represent some sort of discrete functional unit. This argument is quantitatively estimated by the branch length in the mean similarity dendrogram and the corresponding statistical significance.

Back to article page