Acessibilidade / Reportar erro

Estimation of genetic parameters for test day milk records of first lactation Gyr cows using repeatability and random regression animal models

Data comprising 8,183 test day records of 1,273 first lactations of Gyr cows from herds supervised by ABCZ were used to estimate variance components and genetic parameters for milk yield using repeatability and random regression animal models by REML. Genetic modelling of logarithmic (FAS), exponential (FW) curves was compared to orthogonal Legendre polynomials (LP) of order 3 to 5. Residual variance was assumed to be constant in all (ME=1) or some periods of lactation (ME=4). Lactation milk yield in 305-d was also adjusted by an animal model. Genetic variance, heritability and repeatability for test day milk yields estimated by a repeatability animal model were 1.74 kg2, 0.27, and 0.76, respectively. Genetic variance and heritability estimates for lactation milk yield were respectively 121,094.6 and 0.22. Heritability estimates from FAS and FW, respectively, decreased from 0,59 and 0.74 at the beginning of lactation to 0.20 at the end of the period. Except for a fifth-order LP with ME=1, heritability estimates decreased from around 0,70 at early lactation to 0,30 at the end of lactation. Residual variance estimates were slightly smaller for logarithimic than for exponential curves both for homogeneous and heterogeneous variance assumptions. Estimates of residual variance in all stages of lactation decreased as the order of LP increased and depended on the assumption about ME. Estimates of genetic and permanent environment variances did not show any trend due to the increase in the order of LP. Genetic correlation estimates between TD were largest for LP and larger for FW than for FAS. Except for FAS, a similar pattern in genetic correlation estimates was observed for all curves decreasing from values close to unity between adjacent TD at early lactation to negative values for TD in the beginning and the end of lactation. The Legendre polynomial of order five and the FAS under ME4 best fitted the data. There is a potential for using random regression to model animal genetic and permanent environmental effects using test day information of Gyr cows.

genetic evaluation; heritability; selection; variance components; zebu


Sociedade Brasileira de Zootecnia Universidade Federal de Viçosa / Departamento de Zootecnia, 36570-900 Viçosa MG Brazil, Tel.: +55 31 3612-4602, +55 31 3612-4612 - Viçosa - MG - Brazil
E-mail: rbz@sbz.org.br