Biology Reference
In-Depth Information
research. However, the size of these metabolic networks
has made the study of their functions difficult, since they
contain hundreds or thousands of reactions. Thus,
systems-level methods have been developed to gain
insight into how thousands of cellular components func-
tion together as a whole to provide the phenotypes we
observe in living organisms.
experts
databases
Disseminate
Model
Metabolic Reconstructions
Metabolic reconstructions are knowledge bases containing
detailed and organized descriptive information for each
enzyme in an organism. This information includes the
stoichiometry of substrates and products of the reaction it
catalyzes, reaction reversibility, and reaction localization
[3] . This information can readily be found in detailed
biochemical studies and inferred from annotated genomes.
After it is organized in a knowledge base, it can be repre-
sented as a stoichiometric matrix (which we call here the
M-matrix, for metabolism), in which each reaction corre-
sponds to a column and each metabolite to a row.
Metabolic network reconstructions are a relatively
modern development that arose in our efforts to understand
the biochemical processes underlying cell functionality.
Non-genome-scale reconstructions were first developed in
the early 1980s [4,5] . A desire to understand these networks
on a cell-wide level led to the first genome-scale metabolic
reconstruction in 1999 [6] . Recent developments now allow
the partial automation of the reconstruction process for
genome-scale metabolic networks (e.g., via modelSEED
[7] and the SuBliMinaL toolbox [8] ), yielding an ever-
growing number of metabolic reconstructions [3] . Here, we
present an introduction to how these networks are con-
structed and their potential applications, as well as future
directions in the field.
CURATE
VALIDATE
VALIDATE
Growth Rate
0.739 hr -1
CONVERT
DRAFT
RECONSTRUCTION
VS.
0.621 hr -1
Annotated
Genome
FIGURE 12.1 Overview of the reconstruction process. Curation,
conversion, and validation are carried out iteratively until the model's
predictive capabilities are sufficient for the scope of the reconstruction.
anapleurotic reactions ( Figure 12.2 ). The network can be
obtained from http://bigg.ucsd.edu
Stage 1: Creation of a Draft Reconstruction
The process begins with the creation of a preliminary draft
reconstruction. An annotated genome for the organism of
interest is obtained and queried to identify information
such as genomic coordinates, names of loci, other
accepted names, known functions of gene products, and
known splice isoforms. Many of these items are cata-
logued in databases and tools such as Pathway tools [11] ,
metaSHARK [12] ,KEGG [13] , modelSEED [7] ,and
others [14 e 16] , as demonstrated in Box 12.1 . Thus the
acquisition of data for the draft reconstruction is amenable
to automation [7,8] .
For metabolic reconstructions, genes with metabolic
functions need to be identified. Various approaches are
commonly used, including searching for Enzyme
Commission numbers [17] , metabolic terms (e.g., dehy-
drogenase or kinase), or metabolic Gene Ontology terms
[18] . Finally, reactions for each of the metabolic genes must
be established. Again, multiple databases contain such
information, including KEGG [13] , Reactome [19] , and
BRENDA [20] . The metabolic genes and their associated
reactions are collected and organized in preparation for
stage 2, in which the network is manually curated.
THE RECONSTRUCTION PROCESS
The process for constructing a metabolic reconstruction has
been established ( Figure 12.1 ), and a standardized protocol
has been developed to assure the publication of high-
quality genome-scale reconstructions [9] . Here, we high-
light key steps and provide a simplified example of the
reconstruction
process
for
core Escherichia
coli
metabolism.
E. coli
Core Metabolism
The core metabolic reconstruction of E. coli [10] consists
of 95 reactions that catalyze transformations between 72
metabolites and accounts for glycolysis, the TCA cycle, the
pentose phosphate shunt, a simplified version of oxidative
phosphorylation and nitrogen metabolism, as well as
secondary
pathways
such
as
gluconeogenesis
and
Search WWH ::




Custom Search