Genome-Scale Models of Metabolic and Regulatory Networks - Systems Biology: Networks, Models, and Applications - page 233

Table 8.1 Examples of existing genome-scale metabolic network
reconstructions

Number Number of Number of

Organism of genes metabolites reactions Reference

Escherichia coli 904 625 931 [14]

Saccharomyces cerevisiae 750 646 1149 [19]

Staphylococcus aureus 619 571 640 [22]

Haemophilus influenzae 362 343 488 [20]

Helicobacter pylori 268 340 444 [21]

Human mitochondria N/A 226 186 [24]

Streptomyces coelicolor 711 500 971 [125]

Constraints on Metabolic Network Function
In order to study the properties of reconstructed genome-scale
metabolic networks, mathematical modeling tools have to be applied
to the reconstructions. The constraint-based analysis framework
approaches this modeling task by imposing a series of constraints
restricting allowable metabolic flux space (figure 8.1). For the purpose
of constraint-based analysis of metabolic networks, the reconstructed
network structure is represented in the form of a stoichiometric matrix
S . This matrix has M rows with each row corresponding to a metabo-
lite in the network and N columns with each column corresponding
to a reaction in the network. The nonzero elements in each column are
the stoichiometric coefficients of the metabolites participating in a
particular reaction with negative elements corresponding to substrates
of the reaction and positive elements to the products of the reaction.
In addition to representing all the reactions internal to the metabolic
network, including transport reactions, the stoichiometric matrix also

Figure 8.1 Constraint-based analysis framework applied to metabolic network
reconstructions. Imposing a succession of physicochemical constraints on the
metabolic network limits the allowable phenotypic space. Here only
stoichiometric, reaction directionality, and maximum reaction rate constraints
are illustrated. Further constraints due to, for example, transcriptional
regulation can also be imposed on the network to further limit the allowable
phenotypic space.