Protein-protein interaction networks and genome-scale experiments integration to identify disease associated subnetworks

Abstract

It has become clear that the cell is a complex system that can not be described by the action of their isolated components. To study a cell state it is necessary not only to focus on its individual components but also on its complex network of interactions. Methodologies taking into account protein-protein interaction (PPI) relationships have demonstrated that proteins related to a disease tend to be highly interconnected and clustered forming a phenotype subnetworks. Following this network-centric perspective of complex diseases, the main challenge of this Master's Thesis is to develop a methodology capable of integrating high-throughput data, without the necessity of pre-selecting genes of interest, with a PPIs network model in order to identify subnetworks that offer a mechanistic hypotheses underling a disease phenotype. For this purpose, a PPI cell model has been build and a methodology for identifying subnetworks associated to a highthroughput experiment has been designed. Finally, the methodology has been applied to extract and characterize essential subnetworks in cancer cell lines.