The procedures for the bottom-up reconstruction of metabolic networks are well developed. While it is a major and universal cellular process, metabolism is only one example of a cellular function for which we would like to have a genome-scale reconstruction. The established reconstruction procedures for metabolism are now being expanded and adapted for other biochemical reaction networks. Although these efforts are in their early stages, we include in this chapter a description of their status because they are indicative of what is to come. We first describe proteins and some of their properties and how they can be included in a metabolic reconstruction. Then we describe how the process of transcription and translation can be detailed in chemical, and thus stoichiometric, terms. All reconstructed networks in a target organism can be integrated seamlessly within a stoichiometric framework.

Protein Properties

Metabolic network reconstructions describe the chemical transformations that take place among the metabolites in a metabolic network. These reactions occur on the surface of a protein (Figure 7.1). Thus, if properties of proteins can be included in the way a reaction is carried out, then such properties can be included in a reconstruction. If these properties can be stated mathematically then they can be included in a computational model of the network reconstruction.

The chemistry that takes place on the surface of a protein is influenced by many characteristics of the protein molecule. There are many properties of enzymes (Table 7.1) that would be desirable to include in a network reconstruction. Some properties are already built into the GPRs. These would include the specificity versus promiscuity of an enzyme.

Enzyme specificity and efficiency The need for enzyme efficiency and activity can thus be assessed using a network model. During the early stages of evolution, enzymes are expected to have exhibited broad substrate specificity and low catalytic efficiency.