|Bormann, J - GEMBLOUX AGRIC UNIV|
|Druet, T - GEMBLOUX AGRIC UNIV|
|Gengler, N - GEMBLOUX AGRIC UNIV|
Submitted to: Journal of Dairy Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: November 21, 2001
Publication Date: N/A
Interpretive Summary: To improve the accuracy of genetic evaluations of milk, fat, and protein yields, lactation records were calculated from adjusted test-day data. The test-day data were adjusted to remove the environmental effect of the herd on the date when the test-day data were collected and the effects of age, lactation stage, and pregnancy in the current and previous lactations. Data first were adjusted for lactation stage, age, and pregnancy effects. Then estimated breeding value was subtracted to prevent differing genetic merits of cows to bias the estimate of the herd test-day effect. For each herd, the adjusted data was analyzed to estimate the herd test-day effects using a model that included within and across parity effects for each cow. The herd test-day effects then were subtracted from the data without the breeding value adjustment to generate the adjusted test-day data for calculation of lactation records. These lactation records reflect the improved accounting of environmental effects from a test day model and can be combined with historical lactation records. Heritabilities and repeatabilities of lactation records that were adjusted for test day effects were higher than for current lactation records. The adjusted records should improve the accuracy of evaluations and allow the use of test-day data as well as of historical data where test day data are not available.
Technical Abstract: Lactation records for milk, fat, and protein yields were calculated from test day data adjusted for the effects of lactation stage, age, previous days open, days pregnant, and test day class (herd, test date, and milking frequency). These lactation records reflect the improved accounting of environmental effects from a test day model and can be combined with historical lactation records. Test day data were adjusted with existing lactation multiplicative adjustments to maintain variance characteristics. Then, additive adjustments for lactation stage, age, previous days open, and days pregnant were applied. The current multiplicative adjustments for previous days open were not applied because its effect was expected to differ by lactation stage. To remove genetic differences, the estimated breeding value from the previous evaluation divided by 305 was subtracted. Effects of test day class, and permanent environment within and across parities were estimated within herd. The effect of test day class was subtracted from adjusted test day yield, and the breeding value restored. Those deviations then were combined with the best prediction procedure into a lactation measure. Heritabilities and repeatabilities of lactation records that were adjusted for test day class were higher than for current lactation records. The adjusted records should improve the accuracy of evaluations and allow the use of test day data as well as provide for the continued use of historical data when test day data are not available.