Prioritisation and network analysis of Crohn's disease susceptibility genes.

Daniele Muraro,Alison Simmons,Douglas A Lauffenburger,David L Boone

doi:10.1371/journal.pone.0108624

Abstract

Recent Genome-Wide Association Studies (GWAS) have revealed numerous Crohn's disease susceptibility genes and a key challenge now is in understanding how risk polymorphisms in associated genes might contribute to development of this disease. For a gene to contribute to disease phenotype, its risk variant will likely adversely communicate with a variety of other gene products to result in dysregulation of common signaling pathways. A vital challenge is to elucidate pathways of potentially greatest influence on pathological behaviour, in a manner recognizing how multiple relevant genes may yield integrative effect. In this work we apply mathematical analysis of networks involving the list of recently described Crohn's susceptibility genes, to prioritise pathways in relation to their potential development of this disease. Prioritisation was performed by applying a text mining and a diffusion based method (GRAIL, GPEC). Prospective biological significance of the resulting prioritised list of proteins is highlighted by changes in their gene expression levels in Crohn's patients intestinal tissue in comparison with healthy donors.

Highlights

Biological functions are rarely a consequence of the activity of a single molecule and arise from the interactions between multiple components of biological systems
Other pairwise methods analyse relatedness between two genes by applying text mining and assessing a score to the association depending on the degree of similarity in the text describing them within article abstracts [10]
In this work we have prioritised a list of genes associated with Crohn’s disease and developed a graph-theoretical analysis of the molecular interaction network resulting from this list

Summary

Introduction

Biological functions are rarely a consequence of the activity of a single molecule and arise from the interactions between multiple components of biological systems. Since the completion of the human genome project in 2003, high-throughput techniques have generated a large amount of molecular-interaction data in the human cells. The need to analyse the role of associated interaction networks at a system-wide level, rather than focusing on single interactions, led to a change in perspective in the investigation of biological systems and to the development of Systems Biology approaches [1]. Incompleteness in knowledge should suggest caution as these networks are a proxy of the actual interactome, integration with independent functional data may support the biological viability of their topology [4]

Methods

Results

Conclusion