N6-methyladenosine (m6A) plays a critical role in regulating various aspects of mRNA metabolism and translation in eukaryotes. Despite rapid progress in this field, gaps remain in genomics analysis of the m6A epitranscriptome. For example, little is known about how DNA sequence variations may affect the m6A modification and the role of m6A in common diseases. Besides, a computational method to analyze m6A-seq data for differential methylation loci that is compatible with complex study design is lacking. In this thesis, we report two major endeavors to answer these questions. First, we mapped Quantitative Trait Loci (QTL) of m6A peaks in 60 Yoruba lymphoblastoid cell lines. By analyzing these variants, we uncovered features associated with m6A installation, including binding by specific RNA binding proteins (RBPs), RNA secondary structure, and transcriptional processes. Our joint analysis of QTL data of m6A and related molecular traits suggests that the downstream effects of m6A are heterogeneous and context-dependent. We identified new proteins that suppress translation of m6A-modified transcripts. Integrated analysis with GWAS data shows that m6A{QTLs are enriched with variants associated with a range of immune and blood related traits, and contribute significantly to the heritability of these traits. Second, we developed RADAR, a comprehensive analytical tool for detecting differentially methylated loci in MeRIP-seq data. RADAR enables accurate identification of altered methylation sites by accommodating variability of pre-immunoprecipitation expression level and post-immunoprecipitation count using different strategies. In addition, it is compatible with complex study design when covariates need to be incorporated in the analysis. Through simulation and real datasets analyses, we show that RADAR leads to more accurate and reproducible differential methylation analysis results than alternatives.




Downloads Statistics

Download Full History