Variance components, heritability estimates, and breeding values for performance test traits in Old Kladruber horses

Abstract

The Old Kladruber horse is an important Czech genetic resource. In the current study, two categories of traits were evaluated - the first, a numerical score for Type and Gender Expression and the second, 11 traits describing performance divided into four categories: (1) Rideability (Overall Impression, Rideability), (2) Gaits (Walk, Trot, Canter), (3) Carriage Drivability (Dressage Test, Obstacle Driving Test, Marathon Test), and (4) Reliability in Tug (First Tug, Second Tug, Third Tug). The original data set contained records from 700 individuals from the period 1995-2014, each horse having 3-5 performance evaluations for the suite of traits. Our objective was to identify a suitable model for the estimation of genetic parameters and prediction of breeding values. Only one model was examined for the Type and Gender Expression trait, whereas three models were compared for the analysis of each performance trait. Criteria for choosing the most appropriate model were minimal values for the deviance information criterion (DIC) statistics, low ratios of residual variance to phenotypic variance, and maximal heritability estimates. The heritability estimate for Type and Gender Expression was 0.18. For the performance traits, the model with the fewest sources of variation (model 1) was more appropriate than two alternatives with more sources of variation. Heritability estimates from this model ranged from 0.08 to 0.40, while estimates for various performance traits from the other models were in the range 0.08-0.24. Low heritabilities for several of the performance traits suggested that selection for their genetic improvement would likely be unsuccessful. In order to maintain genetic variability, inbreeding, and fitness in the Old Kladruber population we suggest to use breeding value estimation using the heritabilities and the method presented in this paper.

Keywords

• harness horse
• genetic parameters
• performance test
• genetic resource