iConcept Press Logo
Email Password Remember me
Genomics II: Bacteria, Viruses and Metabolic Pathways
Genomics II: Bacteria, Viruses and Metabolic Pathways
iConcept Press

Chapter 3

Genomics II: Bacteria, Viruses and Metabolic Pathways

Pathway-based and Network-based Analysis using Functional Interaction Network

by Irina Kalatskaya and Guanming Wu

Viewed: 2034


A large number of research projects have been recently launched to elucidate the genomic changes present in different forms of cancers (ICGC, TCGA), rare genetic diseases (FORGE in Canada, UK-based programme called Dicephering Developmental Disorders), diabetes (DGAP - Diabetes Genome Anatomy Project) and some others. These projects represent international efforts to improve our comprehension of the molecular basis of different diseases through the application of genome analysis technologies, including large-scale genome sequencing, determination of the gene expression and its methylation status, some others. Overwhelming amount of data generated by these efforts has to be stored, normalized, classified and analyzed. The final and the most challenging task is to extract useful biological or clinically applicable information. It is well know that proteins have to interact to function. Majority of the disease-dependent genomic alterations manifest themselves in the modified or disrupted interactions with their partners and/or in acquisition of new interactive partners. Keeping this in mind we believe that pathway- and network-based analysis of the genomic data is an essential step in the retrieving biological information and our progress in general. The chapter will begin with a brief overview of biological networks, graph properties of biological networks, and network clustering. Next we will describe the Functional Interaction (FI) network: a highly reliable, manually curated pathway-based protein network covering close to 50% of human proteins. We will introduce Reactome FI cytoscape plugin that helps to access the network. This plugin is able to construct a sub-network based on genes of interest, query the FI data source and annotation, retrieve network modules of highly-interacting groups of genes, perform pathway and GO enrichment analysis for the whole network and modules separately, display pathway diagrams, and fetch cancer gene index annotations. And finally we will describe several real life examples on the retrieving useful biological information from high-throughput studies using Reactome FI plugin. First example will show how to generate testable hypothesis based on a set of genes of interest. We will start from set of genes that were differentially expressed between the tumor samples of the patients with esophageal adenocarcinoma containing amplification in 7q21 and associated with poor survival and those with no alteration at this genomic region. Pathway and network-based analysis leads us to the hypothesis that tumors with amplification event produce less T-cell chemoattractants, have fewer tumor-infiltrating lymphocytes and subsequently, these patients have worse survival [Ismail et al., Clin Cancer Research 2011]. FI network might be used not only to generate but also to confirm the hypothesis that was previously generated in laboratory. For example, it was shown that activation of TNF-dependent NF-κB signaling contributes in taxane resistance in tumor cells [Sprowl et al, Breast Cancer Research 2012]. Using pathway and network-based analysis we found a cluster of genes highly enriched in NFKB- and TNF-related signaling pathways. Identical approach was also applied to another two independent datasets that documents differences in gene expression between docetaxel-resistant and parental cancer cell lines with similar results. Other examples will show how to compare tumorigenicity of two sets of genes using FI network [Sawey et al., Cancer Cell 2011] and network clustering analysis for the mutated genes from different cancer data set [Wu et al., Genome Biology 2010].

Author Details

Irina Kalatskaya
Informatics and Bio-computing Platform, Ontario Institute for Cancer Research, Canada, Canada
Guanming Wu
Informatics and Bio-computing Platform, Ontario Institute for Cancer Research, Canada, Canada


Irina Kalatskaya and Guanming Wu. Pathway-based and Network-based Analysis using Functional Interaction Network. In Genomics II: Bacteria, Viruses and Metabolic Pathways. ISBN:978-1-480254-145. iConcept Press. 0000.