Thesis Type: Postgraduate
Institution Of The Thesis: Orta Doğu Teknik Üniversitesi, Graduate School of Informatics, Medical Informatics, Turkey
Approval Date: 2016
Student: ATEFEH LAFZI
Supervisor: YEŞİM AYDIN SONAbstract:
Most of the studies on cancer have tried to explain the observed differential gene expression considering only transcriptional regulation. However, post-transcriptional regulation (PTR) has been increasingly recognized as a complex mechanism that also controls various steps of gene expression regulation. Post-transcritional regulation is governed by the interactions of RNA-binding proteins (RBPs) and microRNAs (miRNAs) with their target genes. In this thesis, having found that several RBPs are differentially expressed in Lung squamous cell carcinoma (LUSC), we developed a statistical model which incorporates copy number variation, DNA Methylation and the regulatory effects of transcription factors, miRNAs and RBPs to predict gene expression in cancer. Including RBP-based regulation in addition to other features significantly increased the Spearman rank correlation between predicted and measured expression of held-out genes. Using a feature selection procedure we identified the candidate RBP regulators in LUSC and confirmed that many of them are also differentially expressed. We also determined the targets of these RBPs and compared them with CLIP-determined targets. Lastly, we performed Kaplan-Meier survival analysis, and showed that some of our candidate RBP regulators have prognostic power in LUSC. Our results suggest that the regulatory effects of RBPs have to be considered to explain differential gene expression in cancer.