| The advancements in the Next-Generation Sequencing (NGS) technology have enabled researchers to study complex genomic questions at the whole genome level. At the same time, the unprecedented massive data generated from NGS experiments poses serious challenges in data analysis and interpretation. One of the most intriguing questions in genomics is how gene transcription is regulated by different elements, such as transcription factors (TFs), histone modifications and DNA methylation. My research primarily focused on analyzing, interpreting and integrating different NGS data sets to understand transcriptional regulation by distinct genomic elements and the associated biological significances in various scenarios.;The first chapter briefly overviews the molecular mechanisms of transcriptional regulation by TFs, histone modifications and DNA methylation.;In the second chapter, we address the problem of peak detection in ChIP-seq experiments for TFs. Numerous algorithms have been developed to detect TF binding sites from ChIP-seq data. However all of these algorithms have several shortcomings, including requirement for matched negative controls and susceptibility to false positives (artifacts). We present Ritornello, a novel control-free and high-fidelity peak calling algorithm we developed for TF binding sites detection.;In the third chapter, we study the role of BCOR as a subunit of a non-canonical PRC1 in repressing the transcription of genes associated with differentiation in human Embryonic Stem Cells (hESCs). We identify the direct gene targets repressed by BCOR and establish the requirement for BCOR in hESC maintenance.;In the fourth chapter, we characterize aging-related alterations in the DNA methylome of mouse hippocampus and investigate their impact on the mouse hippocampal transcriptome and cognitive function. We also discuss dietary restriction's role in preventing aging-related alterations in the DNA methylome and ameliorating cognitive deficits. |
