coherent or in-coherent depending on the parity of negative links in
the loop. FFL is coherent (Fig.5a) if the parity is even and it is
in-coherent (Fig.5b) when the parity is odd.
As sets of genes involved in certain phenotypes are highly
interconnected and regulate each other through coherent and
incoherent regulatory loops (motifs) from different pathways, the
analysis of these can provide insights into the structure and dynam-
ics of the network [5, 35] followed by the identification of disease
biomarkers [36, 37].
The other important network topological parameter of a node
is betweenness centrality, which indicates the sum of the number of
shortest paths from all vertices to all others pass through that node.
Node with high BC serves as a gate keeper in the communication
between different components in a network, for example in Fig. 6
node “A” connects the left and right parts of a network, so it gets
a highest BC value [38] which is a non-intuitive behavior and
plays a crucial role in controlling the dynamics of a system
[4, 34]. Node properties have a significant role in the network
topology [5, 27–29]. For example, networks with degree (number
of edges connected to a node) distribution follow a power law
P(ND) ~ (ND)r, whereris an approximated parameter whose
value ranges in 2r<3, called scale-free networks. There are two
important characteristics of scale-free networks; first they contain
“hubs,” nodes comprising many more connections than others;
second due to hubs, these networks are heterogeneous in terms
of node degree and considered to be robust against single-random
perturbation.
XYZXYZX Y Z X Y ZXYZXYZXYZXYZFig. 5Representation of all possible feedforward loops (either coherent or
in-coherent) in a three-node network
Integrative Workflow for Predicting Disease Signatures 255