and significantly correlated gene, miRNA and TF pairs from the
patient data. Using the workflow as shown in Fig.1, we first
generated gene/miRNA/transcription factor regulatory networks
for primary and metastatic stages of prostate cancer and identified
key regulatory interactions responsible for the transitions from
primary to metastatic tumor stage by integrating patient-derived
gene and microRNA expression data. Analysis carried out by the
workflow revealed thatHOXD10,BCL2,andPGRare the most
important factors affected in primary prostate samples, whereas, in
the metastatic state,STAT3,JUN,andJUNBare playing a central
role. The identified key elements of each network are validated
using patient survival analysis. Our integrative analyses on the
disease network also suggest that some of these molecules are
targeted by differentially expressed miRNAs which may have a
major effect on the dysregulation that led to the disease progres-
sion. We observed that in metastatic prostate tumor, five miRNAs
(miR-671-5p,miR-665,miR-663,miR-512-3p,andmiR-371-5p)
are mainly responsible for the dysregulation ofSTAT3,an impor-
tant player in the tumor metastasis. These observations provide an
opportunity for early detection of metastasis and development of
alternative therapeutic approaches.
Ultimately, the integrative workflow discussed in this chapter
supports deciphering mechanisms underlying complex diseases. As
such, it cannot provide an exact representation of cellular events but
nevertheless guides the formation of hypotheses and their valida-
tion in experiments.
The specific objectives of this chapter are:
l Review of the existing tools and approaches for the analyses of
large-scale biochemical networks.
l Construction of prostate cancer network.
l Integrative workflow to identify tumor-specificsignatures.
l Validation of the identified signatures through experiments/
clinical data.2 Material and Methods
2.1 The Systems
Biology Approach
Biological processes are complex, involving a large variety of com-
ponents that interact in a nonlinear fashion in space and time. The
systems biology approach combines experiments with computa-
tional tools and methods to understand such complex processes
[10, 11]. We consider the systems biology approach as an interdis-
ciplinary collaboration that realizes an iterative cycle of data-driven
modeling and model-driven experimentation (Fig.2). A research
project taking a systems biology approach often starts by gathering
information about a process from the literature and databases. ThisIntegrative Workflow for Predicting Disease Signatures 249