Genome-Scale Models of Metabolic and Regulatory Networks - Systems Biology: Networks, Models, and Applications - page 247

A number of different constraint-based analysis approaches have
been used for strain engineering. Elementary flux mode calculations
were applied to a stoichiometric model of a recombinant yeast strain
[71] in order to analyze effects of knockouts and altered culturing
conditions on poly-b-hydroxybutyrate (PBH) yield. However, the
application of the pathway-based approach is hindered by the compu-
tational complexity of enumerating elementary flux modes for
genome-scale models as well as the need to guess ahead of time which
genetic modifications should be tested. The MoMA approach for
simulating flux changes in response to gene deletions in a genome-
scale metabolic model [38] was used to identify single and multiple
gene deletions that would improve lycopene yield in E. coli [72]. When
the model-based strain design strategy was combined with a transpo-
son insertion-based combinatorial design approach [73], significant
further improvements in lycopene production were achieved.
To complement the model-based metabolic engineering approaches
described above, an efficient in silico method for automatically design-
ing genetically modified strains for metabolite overproduction has
been introduced [74,75]. This method, named OptKnock, attempts to
identify the best possible gene deletions to couple the production of a
desired metabolite to biomass production. The idea is that when a
strain designed by OptKnock is evolved experimentally under a suit-
able selection pressure, increased metabolite production would result
as a side effect of the growth rate increase. The OptKnock approach
uses bilevel optimization where the inner problem is the standard FBA
problem (5) and the outer level is a combinatorial optimization prob-
lem over all possible metabolic network structures with the objective of
maximizing the secretion of a particular metabolite. OptKnock has
been used to identify nonobvious multiple gene knockout strategies for
the production of intermediate metabolites (succinate and lactate) as
well as downstream metabolites (1,3-propanediol, chorismate, alanine,
serine, aspartate, and glutamate) [74,76]. The same basic approach
used in OptKnock has also been extended to designing optimal gene
addition strategies and media compositions for metabolite overpro-
duction [75]. Initial experimental validation of designed lactate
production strains has shown that the OptKnock approach has great
promise in designing strains for metabolic engineering [126].
TRANSCRIPTIONAL REGULATORY NETWORKS
Reconstruction of Regulatory Networks
Transcriptional regulatory networks play a key role in cellular
response to different environments. Recent developments in experimental
techniques have allowed the generation of vast amounts of gene