Network analysis applications in RNA-seq Data = Εφαρμογές ανάλυσης δικτύων σε δεδομένα αλληλούχισης νέας γενιάς.

Περίληψη

Next Generation Sequencing has created a huge amount of data - data that has internal dependencies and interactions. There are currently many tools that allow the primary analysis of NGS data. In this diploma thesis, a tool in R was constructed which allow: (a) the identification of correlations between different genes in transcriptional data, and (b) the analysis of differences in protein interaction networks of human disease models. The polyglutamine (polyQ) neurodegenerative disease spinocerebellar ataxia type 1 (SCA1) is a lethal and progressive disorder caused by CAG expansions in the ataxin-1 (ATXN1) gene. Mutant ATXN1 containing more than 39 CAG repeats encodes the production of a pathogenic protein with an abnormal 3-dimensional conformation. The misfolded protein forms inclusions within the nuclei of neurons and sequesters other nuclear proteins, as well. As a result, proteins in the inclusions, including ataxin-1, lose their normal function, an event that causes cytotoxicity and leads to cell necrosis. Here, we aim in the identification of disease modules within protein interaction networks and molecular mechanisms of dysfunctions that are related to SCA1 progression. To this end, we analyzed RNA-seq data from a cell and a mouse model of SCA1 at three discrete time points of protein aggregation  and compared them with similar data from the cerebellum of a SCA1 patient containing polyQ inclusions. We show that the pathways protein digestion and absorption, ECM-receptor interaction (cells-mice) and PI3K-Akt (cells-mice-patient) signaling are commonly  dysregulated in all datasets.