Using of animal model with genetic grouping for the analysis of some reproductive traits of Iranian Holstein cows

The estimate variance components with a satisfying accuracy of important economic traits such as reproductive traits is a prerequisite for designing breeding strategies. Genetic grouping was suggested in order to predict breeding values of phenotypic records in different years with high accuracy. As there is unknown information in the pedigree of Holstein dairy cows in Iran, genetic group animal models with unknown parents seem necessary. Therefore, the present study was conducted to estimate the genetic parameters, genetic trend and accuracy of estimated breeding values of some reproductive traits (i.e. calving to first service (CTFS), first service to conception (FSTC) and calving interval (CI)) in dairy cows with considering genetic grouping for animals with unknown parents.

Information on calving and insemination dates of the first three calving periods from 3361 herds of the Iranian Holstein, collected by the Animal Breeding Center of Iran during 1981 to 2013 was used. Animals with unknown parents were grouped based on the year of birth and sex and the traits were analyzed using two models, with (model 2) or without genetic grouping (model 1). The model with the lowest Bayesian information criterion (BIC) and the Akaic information criterion (AIC) is considered the best model. Spearman's rank correlation coefficient was used to change in animal rank by considering the genetic grouping. The accuracy of estimated breeding values and genetic trend of the traits was estimated using two models and was compared. Preparation of data for statistical and genetic analysis were carried out using R-software and ASReml software, respectively.

The amount of variance and the standard error of additive variance in model 2 was lower than model 1 and for the residual variance was conversely, but there was no significant difference between the values of the two models. Model 2 was selected the best model for all studied traits based on the goodness of fit criteria. Heritability of CI and CTFS using model 2 was estimated lower (non-significant) than model 1. The heritability value for all reproductive traits was estimated less than 0.05 by two models. The rank of males and females changed duo to genetic grouping. The accuracy of estimated breeding values (EBVs) for all studied traits in model 2 was significantly higher than model 1 (P <0.001). The genetic trend of all traits (exception of the FSTC in the first and third calving period) was positive by model 1 and 2, and the estimated values between the two models were different.

The results of the current study showed that using genetic grouping for reproductive traits analysis of Iranian Holstein cows and accurate prediction of genetic merit of animals seems necessary.