Early Growth and Development and Nonlinear Model Fitting Analysis of Ashidan Yak

**Background:** Ashidan yak is a new hornless breed developed by Chinese scientists using population subculture breeding, based on Qinghai Plateau polled yak parent stock. It has important economic value due to its meat production performance, stress resistance, and reproductive performance. Understanding growth patterns is fundamental for animal genetics and breeding, but no prior research had characterized the growth and development of hornless yak. This study aimed to explore early growth and development of female Ashidan yak and evaluate five nonlinear growth curve models.

**Methods:** The study population comprised 260 healthy female Ashidan yaks living at altitudes exceeding 3200 m, raised under year-round grazing with supplemental feed in cold seasons. Body weight (BW), wither height (WH), body oblique length (BL), and chest girth (CG) were measured at 6, 12, 18, and 30 months of age. Absolute growth rate (G) and relative growth rate (R) were calculated. Five nonlinear models—Logistic, Gompertz, Brody, von Bertalanffy, and Richards—were fitted to the weight data. Model performance was evaluated using mean square error (MSE), Akaike's information criterion (AIC), and Bayesian information criterion (BIC). Statistical analysis was performed using SPSS 16.0 and Excel.

**Key Results:** The coefficient of variation (CV) for body weight ranged from 9.75% to 12.16%, while CV for body size ranged from 4.17% to 7.86%, indicating greater variability in weight than linear measurements. Growth showed marked seasonality: from 6 to 12 months of age (cold season), body weight and chest girth showed negative growth; from 12 to 18 months (warm season), growth peaked; from 18 to 30 months, growth continued but at a reduced rate. Wither height increased steadily across all periods. Among the five models, the Richards model performed best for body weight, with the lowest AIC (4543.98) and BIC (4563.19) values. Estimated mature body weight (parameter A) ranged from 155.462 kg (Richards model) to 6527.970 kg (Brody model). The Brody model produced unrealistically high estimates and deviated substantially from observed growth curves. The Logistic, Gompertz, von Bertalanffy, and Richards models produced similar fitted curves that closely tracked actual measurements, though all overestimated weight at 12 months and underestimated at 6 and 18 months. For body size traits (wither height, body length, chest girth), model fit was poor, with some parameter estimates being biologically implausible.

**Clinical Implications:** This study provides the first characterization of growth patterns in Ashidan yak, demonstrating that their growth deviates from the classic sigmoidal 'S' curve due to strong seasonal environmental influences. The Richards model is recommended for modeling body weight in this breed, though validation with shorter measurement intervals is needed before implementation in breed improvement programs. The finding that growth stagnates or reverses during cold seasons (6–12 months) has direct management implications: supplemental feeding during cold seasons is essential to maintain growth and optimize production. The high phenotypic variability observed (CV 9.75–12.16% for weight) suggests substantial potential for genetic selection. Future work should focus on estimating heritability of growth curve parameters and incorporating them into genomic selection programs to accelerate genetic improvement of this economically important breed.