Abstract

Genetic evaluation for milk, fat and protein yields, using 270-day lactation yields in an across-breed animal model (AM), has been performed in New Zealand since 1996. A new system has been developed to calculate breeding values from herd-test data. In the current AM, a 270-day lactation yield record is calculated based on 1 to 10 test-day records. The environmental effect of herd-test-day (HTD) is removed during the 270-day lactation yield calculation. However, this is done outside the AM, without reference to the genetic level of the HTD contemporary group. A test-day model (TDM) can simultaneously account for the fixed effects such as HTD contemporary groups,and the genetic, permanent environmental (PE) and temporary environmental randomeffects that affect the trait. By including the effect in the model, the TDM allows the environment specific to each test-day to be better taken into account, resulting in an improved accuracy of evaluation. A TDM can accommodate changes in the genetic and PE effects over time by fitting polynomials of time to these effects. A random regression (RR) TDM can accommodate changes in the genetic and PE effects overtime by fitting Legendre polynomials of days in milk (DIM) to these random effects. The RR TDM allows each cow's lactation curve to have a different shape. TDM evaluations are more stable than evaluations obtained using 270-day lactations because they account for genetic differences in maturity rate and persistency of lactation. The sire breeding values (BVs) from the TDM were compared to those from the AM. For sires with more than 30 daughters, the correlations between the BVs from the two models were 0.990, 0.988 and 0.990 for milk, fat and protein yields, respectively. The TDM demonstrated improved stability for sires undergoing their widespread proof. There was less variation in the change in BVs for reproof sires from August through to October compared to the AM. The sires that show the greatest change during early reproof under the AM were sires that were outliers for their TDM persistency breeding values for their breed.

BL, Harris, AM Winkelman, DL Johnson, and WA Montgomerie

Proceedings of the New Zealand Society of Animal Production, Volume 67, Wanaka, 382-387, 2007