Nair, Asha2018-08-142018-08-142018-04https://hdl.handle.net/11299/199034University of Minnesota Ph.D. dissertation.April 2018. Major: Biomedical Informatics and Computational Biology. Advisors: Krishna Kalari, Subbaya Subramanian. 1 computer file (PDF); ix, 116 pages.High-throughput Next Generation RNA sequencing (RNA-Seq) technology is affluent with information about the transcriptome, which includes both protein-coding and multiple non-coding regions. In a diseased state, complex interactions between these regions can go awry. Identification of such interactions is critical to translate the underlying biology of the transcriptome, especially for lethal diseases such as cancer. The field of bioinformatics is currently deficient in workflows that can analyze both coding and non-coding regions together efficiently, to identify disease-specific interactions. In this dissertation, I developed three coherent bioinformatics solutions that aim to address these shortcomings in RNA-Seq. First, a comprehensive workflow called MAPR-Seq was developed to analyze and report various features of protein-coding messenger RNAs. Second, a workflow for non-coding circular RNAs, called Circ-Seq, was developed to identify, quantify and annotate expressed circular RNAs. Third, an integration workflow called ReMIx was developed to identify microRNA response elements (MREs) and integrate them with the different types of RNAs (messenger RNAs, circular RNAs, and microRNAs). Collectively, the three workflows were applied to the largest cohort of breast cancer samples (n=885) from The Cancer Genome Atlas (TCGA). Based on the results obtained from these workflows, I present several key findings that are pertinent to breast cancer. I show that circular RNAs may be a marker for tumor proliferation in estrogen response positive (ER+) breast cancer subtype. I also show how triple negative (TN) breast cancer subtype-specific MRE signatures of messenger RNA – microRNA interactions can be obtained using RNA-Seq data, which has not been explored to date and thus, is a novel undertaking. In the end, my results highlight candidate messenger RNAs, circular RNAs and microRNAs that are found to be associated with MAPK and PI3K/AKT signaling cascades in TN breast cancer subtype. In general, the developed bioinformatics solutions can also be applied to RNA-Seq data of other cancer subtypes and diseases to identify unique messenger RNA – microRNA – circular RNA candidates that could be promising diagnostic targets towards improving treatment options for complex diseases.encircular RNAintegrationmessenger RNAmicro RNAMRERNA-SeqThesis or Dissertation