Leveraging high-throughput datasets for studies of gene regulation

Abstract

In this thesis, I leveraged computational methods on biological data to better understand gene regulation and development of the human body, as well as of the model organisms mouse and yeast. Firstly, I tackled biological questions with machine learning techniques by studying pre-transcriptional gene regulation through nucleosome positioning, which resulted in the identification of function-specific factors and improved predictive performance. Next, computational analysis enabled the discovery of genome-wide epigenetic modifications that play a foundational role in silencing for the monoallelic and monogenic expression of olfactory receptor genes in mice. Lastly, signatures of functional, bound RNA regions provide insight into a potential protocol-specific bias and produce a new avenue for de novo discovery of functional regions.