Random regression models for estimation of covariance functions of growth in Iranian Kurdi sheep

Body weight (BW) records (n=11,659) of 4961 Kurdi sheep from 215 sires and 2085 dams were used to estimate the additive genetic, direct and maternal permanent environmental effects on growth from 1 to 300 days of age. The data were collected from 1993 to 2015 at a breeding station in North Khorasan province; Iran. Genetic parameters for growth traits were estimated using random regression test-day models. The residual variances were modeled by a step function with various classes. The model 16 with a polynomial of 3 order for fixed effect, 6 order for direct genetic effect, 6 order for direct permanent environmental and 6 order for maternal permanent environmental effects with the lowest Bayesian information criterion (BIC) was considered to be the best model. The direct heritability ranged from 0.01 at day 1 of age to 0.36 at 300 days of age. Genetic correlations ranged from 0.03 to 0.98 for body weight between different days of age. The small value of genetic correlation (0.03) among early (day 1) and late (300 days) weights showed that early weights were not under the same genetic control as weights at older ages. Genetic progress was realized and the estimated genetic parameters obtained could be used for further improvement of sheep and small ruminants.