Estimation of genetic parameters of growth curve components obtained from comparison of nonlinear models in Moghani sheep

The traits of economic importance such as body weight at different ages and growth rate in sheep are used as selection criteria in breeding programs. Growth can be defined either as an increase in the number of body cells or an increase in body weight and volume over a period of a lifetime. Animal growth curves can be described using various mathematical models. Growth curve components are heritable in different species. Therefore, the genetic potential of animals can be predicted for growth curve components. The best function describing the growth curve can vary depending on the breed and population under study. The aim of the present study was to investigate the ability of different nonlinear models to fit the growth curve of Moghani sheep and to estimate the genetic parameters of the growth curve characteristics in this breed.

In the current study, the data related to the body weights of Moghani sheep collected at the Moghani sheep breeding station in Jafarabad, Moghan, during the years 1989 – 2016 were used to investigate the growth curve and estimating the genetic parameters of the growth curve characteristics in this breed. Four nonlinear models including Brody, Logistic, Gompertz and von Bertalanffy were used to describe the growth curve of Moghani sheep. All models were fitted on the body weight records of animals at different ages using the NLIN procedure of SAS 9.0 software. The models used have three components including A (maturity weight), B (initial weight of animal) and K (maturity rate). Goodness-of-fit indices including coefficient of determination (R2), Root-Mean-Square Error (RMSE) and Akaike information criterion (AIC) were used to select the best model. After fitting the nonlinear models and selecting the best model, the components of the growth curve of individual animals were estimated using the best nonlinear model. Then, non-genetic factors affecting the growth curve components were investigated using the GLM procedure of SAS. Six univariate animal models that differed in terms of maternal permanent environmental effects and maternal genetic effect as well as covariance between direct and maternal genetic effects were used to analyze the traits. Also, three bivariate analyzes between growth curve components were performed to estimate the genetic correlation between traits.

According to our results, the logistic model with the lowest MSE and AIC and the highest R2 was the best model to describe the growth curve of Moghani sheep. The estimated values for the maturity weight (A), initial weight of animal (B) and maturity rate (K) by the logistic model were 40.328, 7.582 and 0.0270, respectively. The fixed effects of lamb sex, birth type, year and month of calving as well as dam age at calving had a significant effect on the components of the growth curve. Between the six fitted linear models, models number six, two and four were selected as the best models to analyze the components of maturity weight (A), initial weight of animal (B) and maturity rate (K), respectively. Direct heritability for maturity weight (A), initial weight of animal (B) and maturity rate (K) were estimated to 0.17, 0.08 and 0.19, respectively. Estimates of genetic correlation obtained from bivariate analysis between A-B, A-K and B-K were 0•06, 0•03 and −0•0003, respectively.

Among the four nonlinear models used, the logistic model had the best fit for the growth of Moghani sheep. According to the results of our study, the growth curve components of this breed had an acceptable heritability, so that these traits could be used in breeding programs to alter the growth curve and improve the growth pattern of animals in this breed.