Genomic Studies of Reproductive Performance and Fur Quality Traits in American Mink

Abstract

This thesis aimed to enhance reproductive performance and pelt quality traits in American mink (Neogale vison) using genomics approaches. Chapter 3 involved estimation of genetic and phenotypic parameters for pelt quality traits. Low-to-moderate heritabilities (±SE), ranging from 0.12±0.04 to 0.44±0.047, were estimated for dried pelt, live grading, body weight and length traits indicating these traits can be improved by genetic/genomic selection. The estimated genetic correlations demonstrated body weight and length measured in November of the first year of life was a good indicator for pelt size without negative influence on overall quality of dried pelt. The moderate positive genetic correlations between body length in November and harvest with overall quality of dried pelt suggested their utility as indicators to select for increased size and overall quality of dried pelt. Chapter 4 analyzed whole-genome data from 100 mink to detect selection signatures in the genome influencing pelt quality traits and coat color. Selection signatures were detected through three methods of fixation index (Fst), cross population extended haplotype homozygosity (XP-EHH), and nucleotide diversity (θπ). Overlapping top 1% of Fst and XP-EHH contained 376 genes for pelt quality and coat color. Overlapping top 1% of Fst, XP-EHH and θπ revealed 19 selection signature regions on chromosomes 3, 4, 5, 6, 7, 8, 9, and 10, including APCDD1 gene with important roles on hair follicular process. In Chapter 5, genome-wide association studies performed to identify markers associated with eight reproductive traits and five pelt quality traits. The most significant associations were found on chromosomes 1, 2, and 4 for gestation length and on chromosome 6 for dried pelt size. Several candidate genes with important roles in reproduction were detected, along with novel genes related to pelt quality and size. In Chapter 6, prediction performance of three genomic evaluation approaches of genomic best liner unbiased prediction (GBLUP), BayesCπ, and single-step Bayesian multiple marker regression (SSBR) were compared for reproductive and pelt quality traits. SSBR consistently yielded higher predictive accuracy for all traits compared to both GBLUP and BayesCπ. The findings of these studies suggest that genomic approaches hold promise for improving these economically important traits.