Covariance component and genetic parameter estimate of production and fertility traits in Holstein Friesian cross cattle using repeatability animal model

The aim of present study was to estimate expected breeding value (EBV) using repeatability animal model and studying the efficiency of bivariate repeatability model over univariate repeatability model, on the basis of performance records pertaining to fertility and production traits in Holstein Friesian crossbred cattle. Lactation records (5,878) on 1,988 crossbred cows sired by 186 bulls, spread over a period of 34 years (1978 – 2012) were analysed in the study. Estimates of covariance components and genetic parameters for fertility and production traits were obtained using restricted maximum likelihood (REML) approach using average information (AI) algorithm. Estimates of heritability obtained by AIREML were significantly lower in fertility traits in comparison to the production traits. Repeatability model helped in the partitioning of additive, permanent environment and residual variances and thus the upwardly bias due to permanent environment in estimation of additive variance was prevented. The genetic parameter estimates of bivariate repeatability animal model were superior in comparison to the estimates of univariate model. The genetic correlation estimates indicated unfavourable association between fertility and production traits. The bivariate repeatability model had greater potential in identification of sires with higher genetic merit for fertility and production traits.