Abstract:
Bovine milk is an important source of human nutrition in Western countries. My work involved using high-depth sequencing of mRNA, extracted from mammary biopsies of lactating cows, to identify candidate genes that influence lactation. RNA sequence data was used to derive gene expression phenotypes for all annotated genes in the bovine genome, as well as to identify genetic variants in expressed genes. These data were subsequently used to map genetic determinants of gene expression, revealing over 2,700 genes whose expression appears to be influenced by variants that reside largely within non-coding regulatory sequences. To identify the subset of gene expression regulatory variants that might also be influencing lactation, genome-wide association analyses for milk volume and composition were also conducted. Using methods based on correlation, I then observed genetic regions where both gene expression and lactation phenotypes appeared to be co-regulated, highlighting a number of genes that are likely causally involved in impacting lactation processes. Examining the RNA sequence data during variant identification also uncovered evidence of RNA editing in some genes. Systematic examination, after filtering with both computational and manual inspection, yielded over 2,400 sites edited in RNA transcripts. These comprised predominantly (>98%) adenosine-to-inosine edits, the most common RNA edit class in mammals. The proportion of editing was then treated as a phenotype and analysed using similar association methods to those applied for analysis of gene expression and lactation traits. This analysis identified genetic effects for 187 RNA edit sites, some of which showed co-regulation with gene expression and lactation traits, suggesting genetic commonalities between all phenotypes.
