Figure 3

Activation graph of the PID database. (A) The activation graph is built using places from the CADBIOM model as nodes and the dependencies between places as edges. The dependency relationships include the transitions and the influence of conditions on the output of transition. The resulting activation graph contains 9077 nodes and 15499 edges. TGF-β belongs to the major connected component containing 8986 nodes (98% of nodes from the activation graph) demonstrating that all information from PID influences TGF-β signaling. Colors distinct from light blue denote non-connectivity of components. In the zoom view (insert), thin directed edges denote condition dependencies and thick directed edges transition dependencies. (B) Representation of degree distribution. (C) Modularity based hierarchical agglomerative clustering. Colors show the different clusters that are further detailed in Additional file 2: Table S1. A major specific module containing 168 places that include TGF-β, Smad proteins, TGF-β receptors and related molecules such as BMP (members of the TGF-β family) is shown as enlarged red nodes.