Interaction between genotype and geographical region for milk production traits of Iranian Holstein dairy cattle

In order to investigate the presence of genotype by environment (G x E) interactions, first lactation milk production traits of Iranian Holstein cows were used. The data were collected by the National Animal Breeding Center of Iran, Karaj. The records were classified by herds in cold, semi-cold, moderate or warm regions based on climatic data of weather stations. The edited data that were collected between 2004 and 2010, included 618,343, 472,597 and 243,792 test day records for milk, fat and protein yields, respectively. Variance components were estimated using Bayesian's Gibbs sampling method in a random regression model (RRM). For all traits, after examining models according to DIC, better fitting RRM was the model with heterogeneous residual variances. Genetic parameters estimate for milk production traits were different in the four regions. The highest genetic variances were obtained for cold climate regions; and the highest residual variances across the lactation were obtained for warm climate. Genetic variances and heritability estimates revealed unequal genetic expression of Holstein genes under different climates. Generally, for fat and protein yields, the genetic correlation between different regions was lower than that for milk yield, especially for genetic correlations of protein yield between cold and warm regions (ranged from 0.22 to 0.48). The lower correlations between warm and other regions, especially at the beginning and the end of lactation, showed that a trait in different environments was genetically different traits. For milk and fat yields, the high genetic correlations for 305-d yield (> 0.84) between cold, semi-cold and moderate regions indicated a few rank changes of animals over these climates. The results indicated that superior sires in one of regions have their genetic expression for yield limited under other climate conditions and indicated the existence of G x E interaction which may have several implications for the genetic improvement of Holstein cows in Iran. (C) 2014 Elsevier B.V. All rights reserved.