Phenotypic Responses to Selection in Ross 308 Broiler Breeders: Long-Term Growth Assessed With Nonlinear Models...

Phenotypic responses to selection in growth performance of Ross 308 broiler breeders were evaluated using longitudinal body weight records from hatch to 64 weeks across four successive generations (G1-G4). The effects of generation, sex, and their interaction on body weight were examined, and sex-specific growth trajectories were modelled using eight nonlinear growth functions (Gompertz, Richards, Logistic, Brody, von Bertalanffy, Weibull, Hossfeld, and López). Generation and sex exerted highly significant effects on body weight across most ages (p ≤ 0.0001), with males consistently heavier than females and sexual dimorphism becoming evident after 4 weeks of age. At 60 weeks, fourth-generation birds exhibited higher body weights than first-generation birds (males: 5036.5 vs. 4744.3 g; females: 4152.2 vs. 3858.3 g), indicating progressive phenotypic improvement across generations. A significant generation × sex interaction (p ≤ 0.05) indicated that the magnitude of sexual dimorphism varied among generations. Model performance was evaluated using the coefficient of determination (R²), adjusted R² ( $R_{\mathrm{adj}}^2$ ), mean square error (MSE), root mean square error (RMSE), Akaike's Information Criterion (AIC), and Bayesian Information Criterion (BIC). All nonlinear models adequately described the growth patterns of Ross 308 broiler breeders. Among them, the Gompertz function generally provided the best fit across generations and sexes (R² = 0.9978-0.999; $R_{\mathrm{adj}}^2$  = 0.9974-0.999; MSE = 1521.5-1988.5; RMSE = 39.03-44.59), while the Richards model effectively captured sex- and generation-specific variation in growth trajectories. Inflection point weights ranged from 1819.5 to 2012.5 g in males and from 1433.7 to 1573.4 g in females. Predicted values were closely aligned with observed data across models, indicating good descriptive performance of the fitted functions. These findings highlight consistent phenotypic improvement in growth traits across generations and demonstrate that the Gompertz model provides a practical and robust framework for describing and monitoring growth dynamics in broiler breeder populations, supporting its use in selection and management decisions.