Skip to main content
ARS Home » Research » Publications at this Location » Publication #209852
Title: Prediction of Genetic Correlations and International Breeding Values for Missing Traits

Author
item MARK, T - UNIV OF COPENHAGEN
item FIKSE, W - INTERBULL CENTRE
item SULLIVAN, P - CANADIAN DAIRY NETWORK
item Vanraden, Paul

Submitted to: Journal of Dairy Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/5/2007
Publication Date: 10/1/2007
Citation: Mark, T., Fikse, W.F., Sullivan, P.G., Van Raden, P.M. 2007. Prediction of Correlations and International Breeding Values for Missing Traits. Journal of Dairy Science. 90(10):4805-4813.

Interpretive Summary: Prediction of genetic merit for missing traits is possible by combining non-missing indicator traits. The indicator traits were combined by prior genetic correlations obtained from multiple regression equations of estimated genetic correlations among non-missing indicator traits on variables explaining production circumstances and trait definitions. Direct prediction of missing traits was closer to actual breeding values than breeding values for any of the indicator traits. This was verified by pretending that clinical mastitis in each of the Nordic countries was a missing trait. The derived methodology was used to predict breeding values for clinical mastitis in the US for local and international bulls with an average reliability of 43 percent.

Technical Abstract: Prediction of genetic merit for missing traits is possible by combining non-missing indicator traits. The indicator traits were combined by prior genetic correlations obtained from multiple regression equations of estimated genetic correlations among non-missing indicator traits on variables explaining production circumstances and trait definitions. This prediction of missing traits was closer to actual breeding values than breeding values for any of the indicator traits. This was verified by pretending that clinical mastitis in each of the Nordic countries was a missing trait. The derived methodology was used to predict breeding values for clinical mastitis in the US for local and international bulls with an average reliability of 43 percent.