Estimation of the Genetic Parameters for Survival Rate in Lori-Bakhtiari Lambs Using Linear and Weibull Proportional Hazard Models

The data set employed in this study was comprised of a number of 6,800 records of lamb's longevity and their survival rate, collected from 1989 through 2009, from the Lori-Bakhtiari experimental flock at the Shooli Station in Shahrekord, Iran. The data were analyzed using linear models and proportional hazard models with Weibull function. These models included fixed factors and direct additive genetic, maternal additive genetic, maternal common environmental and residual random effects. Genetic parameters were estimated using Restricted Maximum Likelihood (REML) procedure fitting a sire model as well as animal models including different combinations of direct and maternal effects. Estimates of direct heritabilities of longevity and survival rate from different linear models were low (0.01 to 0.09). The maternal heritability ranged from 0.00 to 0.04, and decreaced with increase in the age of lambs. The estimates of heritability in logarithmic scale, original scale and effective heritability obtained from the sire and animal models with Weibull function were medium to high (0.08 to 0.55) and were higher than those estimated through different linear models. The maternal heritability estimated through Weibull models decreased with the age of lambs (0.25 to 0.01). These results indicated that the accuracy of the Estimated Breeding Values (EBV) of lambs' survival rate from birth up to yearling age, using linear animal models vs. proportional hazard models, would be low vs. medium to high, respectively. Both animal and maternal genetic effects should be considered in the models for improving the survival rate up to 4 months of ages.