Genome-Scale Models of Metabolic and Regulatory Networks - Systems Biology: Networks, Models, and Applications - page 248

expression and transcription factor binding data that hold the promise
to speed up network reconstruction significantly. Large-scale regula-
tory network reconstructions can currently be converted to qualitative
in silico models that allow systematic analysis of network behavior
in response to changes in environmental conditions. These models
can be combined with genome-scale metabolic models to build inte-
grated models of cellular function including both metabolism and its
regulation. Since metabolic network reconstruction can be routinely
done for any organism with a sequenced genome, the challenges and
opportunities associated with regulatory network reconstruction are
best understood through a comparison between the two network
types. Some of the key differences between regulatory and metabolic
networks and their respective reconstruction processes are summarized
in table 8.2.

Table 8.2 Some key differences between regulatory and metabolic networks
that affect the network reconstruction process

Metabolic networks Regulatory networks

Structure Fixed reaction Qualitative statements

stoichiometry representing combinatorial

action of different
transcription factors

Evolutionary Enzyme sequences Limited conservation of cis-

conservation highly conserved regulatory sites between

across species closely related species

Malleability Fixed structure in Adjustable structure, because

terms of the substrates of the possiblity that

that a particular enzyme mutations in the cis-

can process regulatory sites change

binding specificity

Level of biochemical Fairly complete Most subnetworks have not

characterization understanding of most been wellcharacterized

subsystems even in model organisms

Modeling Quantitative constraint- Quantitative models can be

approaches based models can be currently constructed only

constructed at the on a small scale; qualitative

genome scale discrete network models

can be used to study large
networks

Role of noise Relatively small, because Possibly significant in

of high concentrations of determining both structural

metabolites involved in features of the network and

most reactions the overall response of the

network to a simulus