A Role for ProteineProtein Interaction Networks in the Identification and Characterization of Potential Biomarkers - Proteomic and Metabolomic Approaches to Biomarker Discovery - page 326

BOX 1

PRINCIPLES OF PROTEIN
NETWORKS
A network is composed of multiple nodes
connected by edges. In protein e protein interac-
tion networks, proteins represent the nodes
(circles in schematics) and the physical protein e
protein interactions represent the edges (lines
connecting the circles in the schematics) in the
network. Furthermore, a node that shows a high
degree of interconnectivity is denoted as a hub
(dark circles). Hub proteins usually have special
functions that are critical for the modular organi-
zation of the interaction network. Second, many
hub proteins interact with each other and
thus from a highly interwoven subnetwork
bridging and regulating different cellular path-
waysandprocesses.Examplesofsomeofthe

well-established hub proteins are p53, Mdm2,
p300, and BRCA1, all of which have been
associated with severe pathological conditions.
Cellular processes are mediated through the
complex action of several biological molecules
through biochemical or biophysical interactions.
Comprehensive understanding of the network
will hence help to better understand the molec-
ular mechanism underlying human diseases.
Proteins

serve as hub proteins
(schematic 1). An alteration (mutation) that
removes one of the hub proteins (schematic 2)
could lead to a signi

“ A ” and “ B ”

cantly more severe outcome
compared to an alteration that affects only
a single edge (schematic 3), which could explain
how different alteration in the same gene could
lead to a different disease phenotype.

of the underlying molecular phenotype. This
need stems from the fact that even though inter-
rogation of DNA (mutations) status or RNA
levels (expression studies) can be informative,
these molecules do not represent the actual
functional unit that carries out the biological
process. Thus the inherent nature of informa-
tion from the protein e protein interaction
networks coupled with other high-throughput
primary data could be bene

NETWORK ANALYSIS USING
PROTEINePROTEIN
INTERACTION

Primary data such as DNA mutations, gene,
and protein expression, and metabolite levels
provides the composition of biological systems;
however, networks such as protein e protein
interaction provide a snapshot of the wiring of
the cellular processes. Thus, interpreting
primary data in context of a cellular network,
such as protein e protein interactions, could

cial when attempt-
ing to classify biological states or

identify

biomarkers.