Determination of the best nonlinear function and genetic parameter estimates of early growth in Romane lambs

The objectives of this study were a) to compare growth functions for describing the early growth curve of Romane sheep based on weighing records, b) to estimate the heritability of the growth curve parameters, and c) to estimate genetic parameters for ۹۰-days-old bodyweight utilizing the data of earlier age. The raw data included ۶۶۲ lambs (۳۱۶ males and ۳۴۶ females) bred at the Romane Sheep Research Center, INRAE, France. The studied trait was the bodyweight of lambs at birth, ۱۵, ۲۱, ۳۵, ۶۰, and ۹۰ days of age. The number of measurements was approximately six for each animal. Dataset after mining consisted of ۳۲۶۱ weight records of ۵۷۴ lambs. We applied four non-linear growth functions, including Gompertz, Brody, Logistic, and Richard. The goodness of fit of the included models were compared using the Akaike information criterion (AIC), coefficient of determination (R۲) and residual mean square (MSE). Predicting abilities of the included models were evaluated by comparing the predicted and observed phenotypes until ۹۰ days of age. Genetic parameters of the non-linear functions were obtained using a specific two steps approach; in first step, the parameters of the different functions were estimated, and in the second, the parameters were considered as observations and we analyzed them using a multiple trait animal models. Residual mean square and R۲ for the models of Brody, Gompertz, Logistic and Richards were ۱۰۶.۷۱ and ۰.۳۷, ۴.۷۹ and ۰.۹۴, ۷.۴۱ and ۰.۸۸, and ۹.۰۴ and ۰.۸۸, respectively. The Logistic function had the smallest AIC and MSE values, and also had the highest R۲ value, indicating the best fit. The estimated heritability of the parameters in the logistic function were low (ranging from ۰.۰۰۷ to ۰.۰۱۷). In our study, the correlation between BV۹۰ and BV۳۵ was ۰.۵۴۱۹ with a confidence interval of ۰.۴۶۹ - ۰.۶۰۸. Since BV۹۰ and BV۳۵ have a positive genetic correlation, BV۳۵ could be used to select the lambs for best growth until the slaughter age of Romane using the Logistic model

Keywords:

meat sheep , Growth curves , nonlinear model , genetic parameter , selection