Document Type : Research Paper

Authors

• Moein Taned ¹
• Mohammad Bagher Zandi ²
• Morad Pasha EskandariNasab ³
• Mohammad Abdoli ⁴

¹ Department of Animal Science, Faculty of Agriculture, University of Zanjan, Zanjan, Iran. E-mail: moein.taned@znu.ac.ir

² Corresponding Author, Department of Animal Science, Faculty of Agriculture, University of Zanjan, Zanjan, Iran. E-mail: mbzandi@znu.ac.ir

³ Department of AnimalScience, Faculty of Agriculture, University of Zanjan, Zanjan, Iran. E-mail: eskandarinasab_M@znu.ac.ir

⁴ Department of Animal Science, Faculty of Agriculture, University of Zanjan, Zanjan, Iran. E-mail: m.abdoli@znu.ac.ir

Abstract

Introduction: Accurate estimation of genetic and phenotypic variance enhances the selection of superior horses and serves as a valuable tool for the long-term improvement of the sport horse population. Therefore, this study was conducted to estimate the variance components for performance traits of sport horses using the Bayesian method.
Materials and Methods: A database was created using 49,026 records from 1499 horses collected between 2017 and 2020 from the Iranian Equestrian Federation. The sport performance traits examined were race completion time (RCT), number of errors in competition (NEC), and rank at the end of the competition (REC). Analysis of variance and Duncan's multiple comparison tests were used to determine the significance of environmental effects, and genetic parameters were estimated using a Gibbs sampling method. R software was utilized to evaluate environmental effects, fit the model, and estimate reliability, variance components, heritability, and genetic correlations, with the GIBBS1F90 and THRGIBBS1F90 software used for estimation. The statistical model included fixed effects for birth year, sex, age, breed, height of obstacles, and level of difficulty of the event, as well as random effects for rider, date, city of competition, and additive genetic effect.
Results and Discussion: Heritability (h2) was estimated using single-trait and multi-trait models for the traits RCT, NEC, and REC, respectively, as 0.02 and 0.08, 0.13 and 0.23, 0.16 and 0.29. The estimated genetic correlations between the traits RCT and NEC, RCT and REC, and NEC and REC were 0.38, 0.36, and 0.65, respectively. The mean estimated reliability (r2) using single-trait and multi-trait models for the traits RCT, NEC, and REC were 0.60 and 0.69, 0.62 and 0.73, 0.58 and 0.66, respectively. The heritability values of different traits can vary, and a specific trait may exhibit different levels of heritability across various populations. The estimated heritability of RCT, NEC, and REC fell within the range of values reported in various horse populations and it was <0.01–0.41, 0.07–0.38, and 0.02–0.23, respectively. These estimates demonstrate genetic variation in the traits within the study population, and their alignment with other studies increases confidence in the estimated values.
Conclusion: The results of the present study indicated that the heritability of the traits studied was low. Among the performance traits, REC showed the highest heritability. Due to its positive genetic correlation with RCT and NEC, selecting for REC could potentially improve the other traits as well. These findings emphasize the importance of using multi-trait models in breeding programs, as they can provide more accurate heritability estimates and enhance the precision and reliability of breeding value predictions.

Keywords

• Bayesian
• Breeding value
• Heritability
• Reliability
• Sport horses