Genome-wide Association Study of Ovary Characteristics in Admixed Beef Heifers Through Comparative Bayesian Approaches

Abstract Reproductive performance of heifers is a key indicator of herd productivity. The objective was to evaluate markers and genes associated with reproductive tract characteristics through a comparative approach. For this study, 295 admixed beef heifers were examined via transrectal ultrasonography at approximately 13 months of age. Exploratory factor analysis of antral follicular count (count per size, side, and total), length, height and diameter for left and right sides of ovaries, and uterine horn diameter revealed a latent variable that included total antral follicle count and ovary diameters per side (n = 3 traits), termed OVARY. Confirmatory factor analysis produced OVARY values per heifer. Quality controls left 116,133 SNP per heifer for association analyses. A marker window-based univariate approach with OVARY was conducted using an empirical Bayesian approach adjusting for nested population structure using marker frequency differences (Method 1) as well as an MCMC Bayesian approach identifying the fraction of SNP to be included with population structure as a fixed effect (Method 2). SNP with high genetic variance in important windows (PPAW > 0.75) were mapped to their closest gene for enrichment analysis. Using Method 1, 2 genes were found to be associated and enrichment analysis identified SMYD2 contributing generally to epigenetic and transcriptional regulatory pathways and GO terms (adjusted P < 0.05). Using Method 2, 11 genes were discovered to be associated with OVARY (adjusted P < 0.05). Three of 11 genes were involved in neural transduction, cornification, epithelial integrity, and adipogenic pathways and GO terms. Eight of 11 genes were involved in reproductive hormonal signaling, post-natal growth, sex determination and reproductive organs development, embryo implantation, and signaling transduction GO terms. Given this, Method 2 provides better insights towards the association of markers and genes for future selection of admixed beef heifers with overall better reproductive characteristics than Method 1.