1- Department of Animal Science, College of Agriculture, Isfahan University of Technology, Isfahan, 841583111, Islamic Republic of Iran.
2- Department of Animal Science, College of Agriculture, Isfahan University of Technology, Isfahan, 841583111, Islamic Republic of Iran. , s.ansari@cc.iut.ac.ir
3- Department of Animal Science, College of Agriculture, Tarbiat Modares University, Tehran, 14115-336, Islamic Republic of Iran.
4- Bioscience and Agriculture Modeling Research Unit, Department of Poultry Science, Tarbiat Modares University, Tehran, 14115-336, Islamic Republic of Iran.
Abstract: (1922 Views)
Four nonlinear models including Logistic, Gompertz-Laird, Richards, and von Bertalanffy were compared to achieve the best prediction of growth parameters describing the growth curve in a crossbred chicken population. Growth data (weekly body weights of chicken from birth to 84 days of age) were collected on 303 birds (174 females and 129 males) of F2 cross of the Arian line broiler chicken (Line B) and Urmia native chicken. Some statistical criteria such as Akaike Information Criterion (AIC), Corrected Akaike Information Criterion for small sample sizes (AICc), and Bayesian Information Criterion (BIC) were used to find the best model. The results showed that the estimated values of the initial weight (W0) and final Weight (Wf) in male were significantly (P< 0.01) higher than the female birds in all models. The average estimated initial weight calculated by Gompertz-Laird (0.038 kg) was closer to the average observed initial weight (0.044 kg). Regardless of sex of the birds, the calculated age (ti) and Weight (Wi) at the inflection point were relatively the same in Gompertz-Laird, Richards and von Bertalanffy models, indicating that the growth patterns described by these models are similar. Meanwhile, the different ti and Wi values between the sexes in the four models revealed the different growth pattern in males and females. The goodness of fit indices (R2 and adjusted R2) were higher than 0.97 in all models, indicating that these models could appropriately be fitted on the growth data. However, based on the AIC, AICc, and BIC criteria, Gompertz-Laird model showed better performance, therefore, it was chosen as the best model to analyze the growth pattern in crossbred of .
Article Type:
Original Research |
Subject:
Insect Physiology Received: 2019/08/19 | Accepted: 2020/09/9 | Published: 2020/10/13