Page 44 - Advances in bioenergy the sustainability challenge

these large, complex data sets. These techniques enable us to develop a systems-level
perspective on cellular activity and an understanding of important contributing networks. 4

As an example, metabolic flux analysis derived from metabolomic data allows us to observe
the flow of material through cellular metabolic pathways. Like a material balance, these fluxes
describe the distribution of material throughout the cell's metabolic network and can help
identify branch points and competing pathways relevant to our desired product. Fluxes also
help to determine the degree of engagement of various enzymes in the pathway, allowing us to

identify rate-limiting steps and control points. 3

Because any biological manipulation will rarely ever produce only an isolated response, it is
important to observe the system-level response of our engineering efforts. Using bioinformatics
and omics technologies allows us to understand the interactions, connections, and responses
between different parts of the system to predict and control the metabolic network.

Designing the Metabolic Network

Once we have sufficient understanding of the organism and its cellular activities, we are then
able to develop and design specific strategies to obtain our desired product. Although we can
introduce, remove, or otherwise modify pathways, identifying the most effective actions a
priori can help save much time and effort. Modern methods to do so are found in the field of
computational systems biology.

A main goal of computational systems biology is to reconstruct cellular networks in silico,
which can model the behavior of the cell. Starting with a cellular model, one is able to
simulate and characterize how possible pathway manipulations will affect the system overall.
Evaluation of these changes can help identify the ideal genetic targets that will maximize our
objectives.

One such method of evaluation is called elementary mode analysis, which uses a systems
engineering approach to decompose metabolic networks into uniquely organized pathways that
can be used to evaluate cellular phenotypes, metabolic network regulation, network robustness,
13
and network fragility. As an extension, neural networks can also be used to make sense of
exceptionally difficult systems and to subsequently predict future behavior. 14

Engineering the Metabolic Network

Once targets and pathways are identified, the next task is to implement these changes in vivo.

This involves genetic manipulation of the host organism using molecular biology. The term
synthetic biology describes the systematic approach to pathway manipulation through
standardized biological components for the purpose of increasing their programmability and
15
robustness. Under this framework, genetic elements are modularized to simplify the process
of genetic engineering. These elements can then be used to introduce new genes, knockout
existing genes, or modify existing deoxyribonucleic acid sequences. Modules can be built up to
produce whole pathways and can also be rearranged to optimize expression.