Skip to main content
ARS Home » Plains Area » Clay Center, Nebraska » U.S. Meat Animal Research Center » Genetics and Animal Breeding » Research » Publications at this Location » Publication #326634
Title: Comparison of SNP and haplotype models for genome-wide association studies for feed efficiency traits in crossbred beef cattle

Author
item SCHWEER, K - University Of Nebraska
item KACHMAN, S - University Of Nebraska
item Kuehn, Larry
item SPANGLER, MATT - University Of Nebraska

Submitted to: International Society for Animal Genetics (ISAG)
Publication Type: Abstract Only
Publication Acceptance Date: 4/22/2016
Publication Date: 7/23/2016
Citation: Schweer, K.R., Kachman, S.D., Kuehn, L.A., Spangler, M. 2016. Comparison of SNP and haplotype models for genome-wide association studies for feed efficiency traits in crossbred beef cattle [abstract]. International Society for Animal Genetics (35th ISAG). Abstract Book. p. 147 (Abstract #P5065). Available: https://www.isag.us.Docs/Proceedings/ISAG_Proceedings_2016.pdf

Interpretive Summary:

Technical Abstract: Feed costs comprise the majority of variable expenses in beef cattle systems making feed efficiency an important economic consideration within the beef industry. Due to the expense of recording feed intake phenotypes, identification of genomic regions associated with feed efficiency traits is advantageous for facilitating selection programs. Genome-wide association studies were performed using 748 crossbred steers and heifers representing seven sire breeds with phenotypes for ADG and DMI. Animals were genotyped with the BovineSNP50v2 BeadChip. Both traits were analyzed independently through SNP (BayesC) and haplotype association studies and together in a bivariate analysis with a haplotype model (BayesIM). In brief, a hidden Markov model (HMM) of variable length haplotype segments is built where haplotypes are mapped to haplotype clusters based on local haplotype similarity. The estimated HMM was then used to assign haplotype cluster genotypes, instead of SNP genotypes, as latent covariates in a Bayesian mixture model. Haplotype cluster effects at loci with non-zero effects were modeled as normal random variables. In the bivariate model, loci where both traits had non-zero effects, cluster effects were modeled as bivariate normal random variables. The number of haplotype clusters at each location was assumed to be either 8 or 16, resulting in a total of three univariate analyses for each trait and two bivariate analyses. Chromosomal regions were defined as 1-Mb windows from the BayesC analyses or 900kb QTL regions produced by the haplotype models. Posterior genomic heritability estimates (SD) for ADG were 0.39 (0.11), 0.42 (0.13), 0.42 (0.13), 0.39 (0.11) and 0.40 (0.15) for BayesC, BayesIM 8 clusters, BayesIM 16 clusters, BayesIM bivariate 8 clusters and BayesIM bivariate 16 clusters, respectively. Dry matter intake posterior genomic heritability estimates (SD) were 0.27 (0.08), 0.35 (0.10), 0.33 (0.10), 0.31 (0.11) and 0.31 (0.12) for the same analyses. Three pleiotropic chromosomal regions in common with all univariate (SNP and haplotype) and bivariate analyses were identified on BTA 1 at 157 Mb, 9 at 4 Mb, and 15 at 68 Mb. These results verify that SNP and haplotype associations yield similar heritability estimates and chromosomal regions.