Design of Pathway-Level Bioprocess Monitoring and Control Strategies Supported by Metabolic Networks - Measurement, Monitoring, Modelling, and Control of Bioprocesses - page 196

elementary modes is mandatory to reduce the complexity of the final model and
facilitate the design of process control.
In this chapter we explore the basic tools to design bioprocess modeling,
monitoring, and control algorithms based on metabolic networks. We start by
reviewing basic properties of metabolic networks, metabolic modeling, and ele-
mentary modes. The envirome layer of information affects critical bioprocess
monitoring and control challenges. The envirome consists of the total quantitative
collection of physicochemical properties that define the extracellular environment.
These are the properties that can be individually or collectively manipulated in a
process and also the ones that are more easily measured in real time. We thus
dedicate a section to the measurement of the envirome. In a recent paper we
explored the possibility of metabolic reconstruction from envirome dynamic data.
We have named this technique ''cell functional enviromics'' [ 7 ]. We show here
how this methodology can be used to design bioprocess control algorithms that
target intracellular control variables such as fluxes or pathways.

2 Genome-Scale Networks Lay the Foundation

In January 2012, the genome online database (GOLD) recorded 3,065 completed
bacterial genome sequences and 7,755 more sequencing projects underway [ 15 ].
Furthermore, the metagenomes (genome of mixed cultures) of 340 sample com-
munities were also recorded in the same database, with 9 % of them from engi-
neered mixed microbial systems (wastewater, solid waste, or bioremediation) [ 16 ].
Genome-scale networks are constructed on the basis of the complete genome
annotation. Identified genes may be associated with metabolic enzymes, mem-
brane transporters, signal transduction, or regulatory control. Combining genome
annotation with basic biochemical information currently available in several dat-
abases (e.g., KEGG [ 17 ] and BioCyc [ 18 ] databases), it is possible to reconstruct
the majority of the metabolic reactions network and also the associated exome-
tabolome [ 19 ]. At least 62 genome-scale metabolic models have been recon-
structed for single organisms, representing 37 genera [ 20 , 21 ], including organisms
of industrial relevance such as Escherichia coli [ 22 ], Saccharomyces cerevisiae
[ 23 ], Pichia pastoris [ 24 , 25 ], and many others. Metabolic networks convey
critical information about the interaction between the extra- and intracellular
phases, which is essential for design of advanced process control strategies that
target intracellular control variables.

2.1 Structure of Metabolic Networks

Studies on the architecture of metabolic networks of microorganisms from the
different domains of life (Eukarya, Bacteria, and Archaea) have shown that cellular
metabolism

has a scale-free topology, which means that most metabolites