Document Type : Research Paper

Authors

• vahid janizadeh ¹
• sadegh alijani ²
• Seyed Abbas Rafat ¹
• Hasan Hasanpour ¹

¹ Department of Animal Sciences, Faculty of Agriculture, Tabriz University, Tabriz, Iran

² Department of Animal Sciences, Faculty of Agriculture, University if Tabriz,, Tabriz,, Iran

Abstract

Keywords

• Ascites
• growth curve
• healthy chicken
• log-logistic model

Main Subjects

• Animal Breeding and Genetics

Extended Abstract

Introduction

Background

The poultry industry is economically vital, driven by increasing socioeconomic demand for animal protein. Given the limitations on expanding livestock and poultry numbers in the country, utilizing animals with high production efficiency is crucial. Diseases can significantly impact a nation’s poultry industry. Contemporary poultry breeding should focus on enhancing genetic resistance to diseases, thereby reducing antimicrobial reliance. Current selection practices for commercial lines with accelerated growth, increased production, and lower feed conversion ratios have inadvertently led to reduced genetic diversity, physiological disorders, and compromised immune system function. This erosion of genetic diversity creates populations vulnerable to emerging diseases, posing a substantial risk to the poultry industry. Ascites, or pulmonary hypertension syndrome, is a metabolic disorder observed in broiler chickens resulting from right ventricular failure accompanied by fluid accumulation in the abdominal cavity. Growth models are valuable management tools for understanding the influence of intrinsic and environmental factors on growth-related traits. This study aimed to evaluate various models for fitting growth curves in healthy and ascites chickens, identifying the best-fitting model using classical statistical methods.

Methods

 To investigate the growth curves of broiler chickens at the line level, body weight records of 823 chicks from the paternal B-line of the commercial Arian strain were used. A total of 381 male and 442 female chicks were raised on litter, with feeding and rearing conditions conforming to the guidelines of the Line Breeding and Improvement Complex in Babolkanar. The study was conducted at the Research Station of the University of Tabriz. A lighting program of 23 hours of light and 1 hour of darkness was employed throughout the rearing period. Birds had ad libitum access to water and feed. The paternal B-line has undergone continuous selection for increased body weight over multiple generations. Chicks were fed starter feed from days 0-15, grower feed from days 15-30, and finisher feed from days 30-45, and were vaccinated according to standard protocols. Body weight records were collected at hatch, weekly from weeks 1 to 6, and at 45 days of age. Following the removal of outlier data from weekly weight measurements, conventional methods were employed to fit various nonlinear mixed models to investigate growth curves in ascetic and healthy chickens.

Results

 The results of parameter estimation with a general random effect for 8 nonlinear mixed models showed that the Mitscherlich model for healthy and ascetic males with an AIC index of 33945 and 7012.50, respectively, and the log-logistic model for healthy and ascetic females with an AIC index of 41061 and 1468.30, respectively, were selected as the best models. These two models, having three parameters (B, A, and K), each considered with a random effect, were examined. The lowest values of the AIC and BIC fit criteria for the log-logistic model with two random parameters (B, K) were 53985 and 54038, respectively, followed by the three random parameter model (B, A, and K) with values of 54069 and 54006, respectively.

Conclusion

     In conclusion, the nonlinear Logistic and Mitscherlich models with a general random effect provided the best fit to the body weight data for describing the growth of healthy and ascetic male and female chicks. Furthermore, significant differences in the growth patterns between these two groups were observed, and these differences were well-described by the nonlinear mixed-effects models.

Data Availability Statement

This article contains all the data that were created or evaluated during the research.

Acknowledgements

The authors would like to sincerely thank the members of the Faculty of Animal Sciences, University of Tehran Research Council for the approval and support of this research.

Conflict of interest

The author declares no conflict of interest.