A meta-analysis of genetic parameter estimates for conformation traits in horses

In the breeding plans of animals, it is essential to have sound estimates of genetic parameters for economically important traits. Over the previous years, genetic parameter estimates have been reported for conformation traits in horses. However, these estimates have been got from studies based on various horse populations. This has resulted in considerable variability among heritability and genetic correlation estimates. Thus, the objective of this study was to conduct a meta-analysis based on a random-effects model to combine different published heritability estimates and genetic correlations between conformation traits (measured and scored) in horses. In total, 166 heritability and 72 genetic correlation estimates from 30 peer-reviewed articles published in 16 different scientific journals between 1995 and 2020 were used in this study. Chi-square (Q) test and the I2 statistic were performed to measure heterogeneity of genetic parameter estimates. Also, Egger's linear regression asymmetry was used to examine the presence of publication bias. The weighted coefficients of variation for the conformation traits varied from 0.27% (for leg stance) to 37.80% (for temperament). The heritability estimates for the measured and scored conformation traits ranged from 0.142 (for neck length) to 0.605 (for height at croup), and from 0.082 (for shoulder angle) to 0.445 (for head), respectively. Based on the values of the I2 index, substantial and or considerable heterogeneities were observed for the heritability estimates of the measured conformation traits. Also, the results of the Q test indicated that all measured conformation traits showed significant heterogeneities (P<0.10). Except for the genetic correlations between height at withers with leg quality, leg stance, and hooves which were low and non-significant, the genetic correlations between height at withers with other conformation traits were high and significant (P 0.05), and ranged from 0.455 (with chest width) to 0.942 (with height at croup). The genetic correlation estimates between cannon bone circumference with other conformation traits were medium to high and varied from 0.382 (with chest width) to 0.688 (with height at croup). The genetic correlations between cannon bone circumference with chest and or thorax circumference, and body length were non-significant (P 0.05), and their 95% confidence interval included zero. Therefore, these correlation estimates would be considered equal to zero. The average genetic parameter estimates reported in this meta-analysis study are applicable in breeding designs when reliable estimates are not available for conformation traits in horse populations.