Submitted to: Journal of Dairy Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/7/2001
Publication Date: N/A
Citation: Interpretive Summary: To improve the accuracy of genetic evaluations of milk, fat, and protein yields, test-day data were analyzed to determine the effect of lactation stage, age, and pregnancy and to estimate variance ratios needed to calculate evaluations. Analysis of test-day data before combining it into lactation records enables better accounting of environmental effects. The estimates from this study can be used to calculate lactation values which can be analyzed with current genetic evaluation programs along with lactation records from before the period when test-day data is available. This enables a transition from a lactation to a test-day system without losing historical data. This analysis used data from Jersey and Holstein cows. For Holsetins, datasets were selected from herds in California, Pennsylvania, Texas, and Wisconsin. Datasets included 0.7 to 7.7 million test-day records. To develop lactation values compatible with historical lactation records the same multiplicative adjustments were applied to the test-day data as are applied to lactation records so that variance characteristics would be similar. Datasets without these adjustments also were analyzed to determine the effect of adjustment. Heritabilities from the unadjusted data ranged from 0.19 to 0.30 for milk, 0.13 to 0.15 for fat, and 0.17 to 0.23 for protein. The adjustments had large effect on the solutions for age as expected, and some effect on lactation stage. The solutions developed from the adjusted data can be used to adjust test-day yields before evaluation analysis. The variance components and solutions for the various effects can be used to calculate test-day deviations in a within-herd analysis that contributes to an across-herd analysis.
Technical Abstract: Test-day variances for permanent environmental effects within and across parity were estimated along with lactation stage, age, and pregnancy effects for use with a test-day model. Data were test-day records for calvings since 1990 for US Jerseys and for US Holsteins from California, Pennsylvania, Texas, and Wisconsin. Single-trait repeatability models were fitted for milk, fat, and protein test-day yields. Method R and a preconditioned conjugate gradient equation solver were used for variance component estimation because of large data sets (0.7 to 7.7 million records). Test-day data were adjusted for environmental effects of age, calving season, and milking frequency and for estimated breeding value (EBV) expressed on a daily basis. Variance ratios (residual divided by variance of effect) within parity were similar across breed, region, and sample: 1.5 to 1.8 for milk, 3.0 to 4.3 for fat, and 1.8 to 2.3 for protein. Variance ratios across parities ranged from 3.5 to 6.8 for milk, 8.7 to 17.6 for fat, and 5.5 to 9.4 for protein. Adjustment for EBV reduced permanent environmental variance across parity as well as removing cow genetic variance. Adjustment factors for lactation stage, age at milking, and pregnancy were estimated from test-day data based on the same single-trait repeatability models and variance ratios estimated for permanent environment within and across parities. Those factors can be used to adjust test-day yields before evaluation analysis. The variance components and solutions for the various effects can be used to calculate test-day deviations in a within-herd analysis that contributes to an across-herd analysis.