|O'Connell, Jeffrey -|
Submitted to: Journal of Dairy Science
Publication Type: Abstract Only
Publication Acceptance Date: March 5, 2012
Publication Date: June 28, 2012
Citation: Wiggans, G.R., Cooper, T.A., Van Raden, P.M., O'Connell, J.R., Bacheller, L.R. 2012. Maternal grandsire confirmation and discovery in dairy cattle. Journal of Dairy Science. 95(Suppl. 2):400(abstr. 323). 2012. Technical Abstract: Accurate pedigree information is essential for selecting dairy animals to improve economically important traits. Two methods of maternal grandsire (MGS) discovery were compared. The first compared one single nucleotide polymorphism (SNP) at a time using a genotype from one or both parents (SNP method). The second compared haplotypes of a potential MGS to the animal’s maternal haplotype, which included linkage information across all loci (HAP method). Modified pedigrees with 5% of true MGS replaced by a random genotyped bull from the same birth year and 5% set to missing were created to test the HAP method. To test the SNP method, the same group of animals (including multiple genotypes for some animals and excluding imputed dams) was used. Both methods ranked the most likely MGS. The SNP method can be performed as soon as a genotype is received because no imputation is required. The HAP method was expected to have greater accuracy because it includes low-density genotypes imputed to the 45,187 SNP used for genomic evaluation as well as additional imputed animals. Accuracy of predicting true MGS was evaluated using genotypes for 4,620 Holsteins, 659 Jerseys, and 160 Brown Swiss with confirmed genotyped sires for the SNP method and modified pedigrees for 4,134 Holsteins, 552 Jerseys, and 142 Brown Swiss for the HAP method. Accuracy of true MGS prediction for Holsteins, Jerseys, and Browns Swiss was 95, 91, and 94%, respectively, with the SNP method and 97, 95, and 97% with the HAP method. Lack of imputed SNP decreased accuracy of the SNP method for low-density genotyped animals (78%) compared with BovineSNP50 genotyped animals (97%). Predicted MGS accuracy with the HAP method was 98% for BovineSNP50 genotypes, 94% for Bovine3K genotypes, and 92% for imputed genotypes. When the HAP method was extended to great grandsires, accuracy of maternal great-grandsire confirmation and discovery was 92% for 652 Holsteins, 95% for 33 Jerseys, and 85% for 20 Brown Swiss. Because most dairy bulls have been genotyped, parentage and MGS analysis can accurately confirm, correct, or discover parents and more distant ancestors for most animals.