Skip to main content
Figure 1 | BMC Systems Biology

Figure 1

From: Network-based scoring system for genome-scale metabolic reconstructions

Figure 1

Illustration of the method. Bipartite network representation and corresponding dendrogram. We illustrate the model with four coupled stoichiometric equations in the pentose-phosphate pathway of E. coli. Reaction acronyms stand for the catalyzing enzyme: zwf, glucose- 6- phosphate dehydrogenase [EC 1. 1.1. 49]; pgl, 6- phospho-gluconolactonase [EC 3. 1. 1.31]; gnd, 6- phosphogluconate dehydrogenase [EC 1.1.1. 43]; rpe, ribulose- phosphate 3- epimerase [EC 5. 1.3. 1]. These equations are represented as an unweighted undirected bipartite network formed by connections (black lines) between reactions (blue squares) and metabolites (orange circles). Notice that metabolites or reactions are not connected among themselves. The model adjusts an underlying binary tree to the observed network structure where each internal node t is associated with a tree probability ρ t , that is transformed into a distance d t = 1 ρ t . Every pair reaction-metabolite with minimum common ancestor t is separated a distance d mr = d t (the NaN notation indicates that only one class of nodes populates the leaves of the child branches such that the corresponding internal node does not contribute).

Back to article page