Comparison of Two Singles- and Multiple Trait Random Regression Models in Estimation of Genetic Parameters of Production Traits in Holstein Dairy Cattle

In this research test-day records of milk, fat and protein yields of first lactation of 11368 Holstein cows were used to comparison of estimated genetic parameters by restricted maximum likelihood method based on single trait and multiple trait random regression models. These data were collected from 133 herds during 2001 to 2010 by Cattle Breeding and Dairy Product Improvement Center. Heritability of milk, fat and protein yield during lactation based on single trait model were ranged from 0.107 to 0.226, 0.049 to 0.145 and 0.087 to 0.239, respectively, but by multiple trait model these values were ranged from 0.109 to 0.223, 0.051 to 0.118 and 0.084 to 0.190, respectively. Minimum genetic and permanent environmental variances were in peak to mid lactation. Assuming independent of traits during lactation, permanent environmental variance was underestimated and heritability was overestimated nearly to end of lactation. Genetic correlations between production traits during lactation were not constant and it was higher at the end of lactation. However, the largest genetic correlation was observed between milk and protein yield.