Computational protocol: Overdominance Effect of the Bovine Ghrelin Receptor (GHSR1a)-DelR242 Locus on Growth in Japanese Shorthorn Weaner Bulls: Heterozygote Advantage in Bull Selection and Molecular Mechanisms

Similar protocols

Protocol publication

[…] A mixed-inherited animal model was used to evaluate the effect of genotype and to compute basic summary statistics on traits of interest. The snp_ad, snp_a, snp_d, and epi_snp options of the Qxpak software () were used for analyzing the additive and dominant effects of the DelR242 [or nt-7(C > A)] locus and for the effect of epistasis between the DelR242 and nt-7(C > A) loci. The following models were used:Exp. 1 yijkno=yeari+areaj+bxk+SNP+uo+eijknoExp. 2 yijklmno=yeari+areaj+bxk+sexl+shippingm+SNP+uo+eijklmno,where yijkno or yijklmno is an individual phenotypic observation; the effect of year i (i, 14 or 3 levels for Exp. 1 or Exp. 2), the grazing area effect (areaj) (j, 5 or 4 levels for Exp. 1 or Exp. 2), the sex effect (sexl) (l, 2 levels for Exp. 2), and the shipping month effect (shippingm) (m, 12 levels for Exp. 2) were used as fixed effects. b is a regression coefficient of age (xk) at the start of direct testing (Exp. 1) or of month of shipping (Exp. 2). In this analysis, the infinitesimal genetic effect was included and treated as a random effect (uo) with covariance matrix, Aσu2 (where A is the numerator relationship matrix). eijkno and eijklmno are vectors of random residual effects. SNP is the single-locus SNP (including the DelR242) genotypic effect, which was partitioned into additive (a) and dominance (d) effects.The likelihood ratio test was performed by removing the single-locus SNP genotypic effect from the model, and normal P values were obtained by assuming a χ2 distribution of the likelihood ratio test (degrees of freedom = 2).The proportion of genotypic variance explained by the model was calculated as follows:Variance ratio=[2pq(a+d(q−p))2]/VAThe ratio of the variance of dominance deviation to the additive genetic variance of the trait was calculated as follows:Variance ratio=[(2pqd)2]/VA,where p and q are allelic frequencies, a and d are additive and dominant effects of the DelR242 locus, respectively, and VA is the additive genetic variance of the trait obtained from an animal model ignoring single-locus SNP genotypic effects ().For least-squares means analysis, we performed analyses of variance (ANOVAs) with the generalized linear model (GLM) procedure in SAS () fitting the fixed effects of genotype (Exp. 1 and Exp. 2), test year (Exp. 1 or Exp. 2), grazing area (Exp. 1 or Exp. 2), sex (Exp. 2), shipping month (Exp. 2), sire (17 levels for Exp. 2), and a linear covariate of age of animals on testing (Exp. 1) or shipping (Exp. 2). The significance threshold was set to a Bonferroni-corrected P depending on the number of traits in Exp. 1. [...] In total, 30 nucleotide variations (26 SNPs, 3 indels including DelR242 and intron 1 microsatellite [(GTTT)5 or (GTTT)6] of 23 individuals (nine Bos taurus breeds; accession no.: AB492155–AB492177) were used to estimate local LD and haplotype blocks for a 5.8-kb genomic region that surrounds the DelR242 locus of GHSR1a. The genotype data [nt-7(C > A), nt456(G > A), and DelR242] of 95 sires and 540 half sibs were also used to estimate local LD. Local LD (r2) and common haplotype patterns were computed with Haploview version 4.2 (). […]

Pipeline specifications

Software tools QxPak, Haploview
Application GWAS
Organisms Bos taurus