Identifying Functional Regulatory Regions in the Bovine Genome

Submission note: A thesis submitted in total fulfilment of the requirements of the degree of Doctor of Philosophy to the School of Applied Systems Biology, College of Science, Health and Engineering, La Trobe University, Victoria, Australia.

Thesis with publications.

Extensive research in both humans and livestock suggests that complex trait variation is typically the result of many polymorphisms, each of small effect, in noncoding (likely regulatory) rather than coding regions. However, pinpointing these causative variants has been difficult, partly due to a limited understanding of functional noncoding sequences. This thesis hypothesized that improving the identification of functional noncoding bovine genome would help prioritise sequence variants, improve assessment of noncoding sequence contribution to genetic variation, and inform genomic prediction in cattle in general. In this thesis the first research endeavour was annotating enhancers in the bovine genome using both a mammalian genomic synteny approach and some bovine chromatin modification assays. Analysis revealed that enhancers identified through bovine-specific experiments were enriched for sequence variants associated with bovine milk production traits, and more so than enhancers identified through sequence homology. In the second research chapter of this thesis, mega-base-size chromatin domains (called ‘TADs’) were mapped to the bovine genome through sequence conservation. QTLs associated with gene expression variation from a dairy cattle dataset were located within the same TAD as their target genes more often than expected by chance, proving the mapped TADs could be used to reduce the search space for expression QTLs. Lastly, microRNAs expressed in 17 bovine tissues, along with their in silico predicted target sites on messenger RNAs, were identified in the bovine genome from libraries of bovine RNA sequence. A considerable number of novel microRNAs had not been previously described. A catalogue of putative microRNA and messenger RNA interacting sites were also characterised. In conclusion, this thesis has progressed genetically informed knowledge of the layers of noncoding regulation, from enhancers to TADs to microRNAs, which will aid the functional annotation of bovine genome. The knowledge created improves the understanding of noncoding sequence contribution to genetic variation and allows a more targeted approach when assessing the pool of sequence variants that have to be investigated for optimal genomic prediction of complex mammalian traits, including those important to the dairy industry.