Page Banner

United States Department of Agriculture

Agricultural Research Service

Title: Studies on the Value of Incorporating the Effect of Dominance in Genetic Evaluations of Dairy Cattle, Beef Cattle, and Swine

Authors
item Van Tassell, Curtis
item Misztal, Ignacy - UNIV OF GEORGIA
item Varona, Luis - UNIV OF GEORGIA
item Culbertson, Matthew - UNIV OF GEORGIA
item Bertrand, J. Keigh - UNIV OF GEORGIA
item Mabry, John - UNIV OF GEORGIA
item Lawlor, Thomas - HOLSTEIN ASSOCIATION
item Gengler, Nicolas - FACULTE UNIVERSIATIRE

Submitted to: Biotechnology, Agronomy, Society, and Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 29, 1998
Publication Date: N/A

Interpretive Summary: The total genetic merit of an animal can be expressed as a combination of the estimated additive breeding value of the animal and the specific combining ability of its parents, sometimes called nicking. The components of nicking are the effects of dominance, epistasis, and inbreeding. Dominance effects are currently ignored in national genetic evaluations of farm animals because of ignorance of the level of variation for those effects for traits of interest and because of difficult computational problems. The potential gains from including effects of dominance in genetic evaluations include better estimates of breeding value and the prediction of the effects of nicking for each pair of prospective parents. This study focused on making evaluation with dominance effects feasible computationally and on ascertaining benefits of such an evaluation for dairy cattle, beef cattle, and swine. Dominance effects were found to account for up to 10% of phenotypic variation; estimates were larger for growth traits. Changes in ranking from ignoring dominance effects were important for dairy cattle but were less important for swine. Predictions of nicking effects could be used in computerized mating programs via the Internet. Gains from including the dominance effect in genetic evaluations would be modest but would outweigh expenditures and would result in more accurate evaluations on which to base breeding decisions.

Technical Abstract: Nonadditive genetic effects are currently ignored in national genetic evaluations of farm animals because of ignorance of the level of dominance variance for traits of interest and the difficult computational problems involved. Potential gains from including effects of dominance in genetic evaluations include "purification" of additive values and availability of predictions of specific combining abilities for each pair of prospective parents. This study focused on making evaluation with dominance effects feasible computationally and on ascertaining benefits of such an evaluation for dairy cattle, beef cattle, and swine. Using iteration on data, computing costs for evaluation with dominance effects included could be less than twice expensive as with only an additive model. With Method R, variance components could be estimated for problems involving up to 10 million equations. Dominance effects accounted for up to 10% of phenotypic variance; estimates were larger for growth traits. As a percentage of additive variance, the estimate of dominance variance reached 78% for 21- day litter weight of swine and 47% for post weaning weight of beef cattle. When dominance effects are ignored, additive evaluations are "contaminated"; effects are greatest for evaluations of dams in a single large family. These changes in ranking were important for dairy cattle, especially for dams of full-sibs, but were less important for swine. Specific combining abilities cannot be included in sire evaluations and need to be computed separately for each set of parents. Predictions of specific combining abilities could be used in computerized mating programs via the Internet. Gains from including the dominance effect in genetic evaluations would be moderate but would outweigh expenditures.

Last Modified: 10/25/2014
Footer Content Back to Top of Page