Invited review: Reliability computation from the animal model era to the single-step genomic model era.

The calculation of exact reliabilities involving the inversion of mixed model equations poses a heavy computational challenge when the system of equations is large. This has prompted the development of different approximation methods. We give an overview of the various methods and computational approaches in calculating reliability from the era before the animal model to the era of single-step genomic models. The different methods are discussed in terms of modeling, development, and applicability in large dairy cattle populations. The paper also describes the problems faced in reliability computation. Many details dispersed throughout the literature are presented in this paper. It is clear that a universal solution applicable to every model and input data may not be possible, but we point out several efficient and accurate algorithms developed recently for a variety of very large genomic evaluations.

Information from later lactations improves accuracy of genomic predictions of fertility-related disorders in Norwegian Red.

Our aim was to investigate whether including information from later lactations improves accuracy of genomic breeding values for 4 fertility-related disorders: cystic ovaries, retained placenta, metritis, and silent heat. Data consisted of health records from 6,015,245 lactations from 2,480,976 Norwegian Red cows, recorded from 1979 to 2012. These were daughters of 3,675 artificial insemination bulls. The mean frequency of these disorders for cows in lactation 1 to 5 ranged from 0.6 to 2.4% for cystic ovaries, 1.0 to 1.5% for metritis, 1.9 to 4.1% for retained placenta, and 2.4 to 3.8% for silent heat. Genomic information was available for all sires, and the 312 youngest bulls were used for validation. After standard editing of a 25K/54K single nucleotide polymorphism data set that was imputed both ways, a total of 48,249 single nucleotide polymorphism loci were available for genomic predictions. Genomic breeding values were predicted using univariate genomic BLUP for the first lactation only and for the first 5 lactations and multivariate genomic BLUP with 5 lactations for each disorder was also used for genomic predictions. Correlations between estimated breeding values for the 4 traits in 5 lactations with predicted genomic breeding values were compared. Accuracy ranged from 0.47 and 0.51 for cystic ovaries, 0.50 to 0.74 for retained placenta, 0.21 to 0.47 for metritis, and 0.22 to 0.60 for silent heat. Including later lactations in a multitrait genomic BLUP improved accuracy of genomic estimated breeding values for cystic ovaries, retained placenta, and silent heat, whereas for metritis no obvious advantage in accuracy was found.

Short communication: genetic parameters for fertility-related disorders in Norwegian Red.

Heritabilities and genetic correlations were estimated for the 4 most common fertility-related disorders in Norwegian Red: retained placenta, cystic ovaries, silent heat, and metritis. Data on 1,747,500 lactations from 780,114 cows calving from January 2001 through December 2011 were analyzed using multivariate threshold sire models to estimate variance components for the 4 disorders in the first 5 lactations. The traits were defined as binary within lactation (0=unaffected, 1=affected), and each fertility-related disorder was analyzed separately with the 5 lactations as correlated traits. The mean frequency of affected cows ranged from 0.5 to 1.7% for cystic ovaries, 0.7 to 1.1% for metritis, 1.3 to 3.4% for retained placenta, and 1.7 to 2.7% for silent heat. Posterior means (standard deviations) of heritability of liability ranged from 0.02 (0.01) to 0.12 (0.01), and were lowest for silent heat and highest for cystic ovaries. Genetic correlations across lactation within disorder were positive and moderate to high, ranging from 0.79 to 0.95 for cystic ovaries, 0.40 to 0.75 for metritis, 0.53 to 0.94 for retained placenta, and 0.39 to 0.83 for silent heat.