Genomic analyses of withers height and linear conformation traits in German Warmblood horses using imputed sequence-level genotypes

Abstract

Background: Body conformation, including withers height, is a major selection criterion in horse breeding and is associated with other important traits, such as health and performance. However, little is known about the genomic background of equine conformation. Therefore, the aim of this study was to use imputed sequence-level genotypes from up to 4891 German Warmblood horses to identify genomic regions associated with withers height and linear conformation traits. Furthermore, the traits were genetically characterised and putative causal variants for withers height were detected.

Results: A genome-wide association study (GWAS) for withers height confirmed the presence of a previously known quantitative trait locus (QTL) on Equus caballus (ECA) chromosome 3 close to the LCORL/NCAPG locus, which explained 16% of the phenotypic variance for withers height. An additional significant association signal was detected on ECA1. Further investigations of the region on ECA3 identified a few promising candidate causal variants for withers height, including a nonsense mutation in the coding sequence of the LCORL gene. The estimated heritability for withers height was 0.53 and ranged from 0 to 0.34 for the conformation traits. GWAS identified significantly associated variants for more than half of the investigated conformation traits, among which 13 showed a peak on ECA3 in the same region as withers height. Genetic parameter estimation revealed high genetic correlations between these traits and withers height for the QTL on ECA3.

Conclusions: The use of imputed sequence-level genotypes from a large study cohort led to the discovery of novel QTL associated with conformation traits in German Warmblood horses. The results indicate the high relevance of the QTL on ECA3 for various conformation traits, including withers height, and contribute to deciphering causal mutations for body size in horses.

Fig. 1

Estimated proportions of the phenotypic variance for withers height. Proportions of the variance explained by each of the 31 equine autosomes (dark blue) and the two quantitative trait loci on chromosomes 1 and 3 (light blue). Summed up, the marker-based heritability for withers height was 0.51

Fig. 2

Results of the genome-wide association study for withers height in 2709 mares. Manhattan plots of the –log₁₀ p-values for the association of variants with withers height for the whole genome (a) and the two quantitative trait loci on Equus caballus chromosome 1 (ECA1) (b) and 3 (ECA3) (c). In d, the top associated SNP rs68603062 from the preliminary GWAS (a) was included in the model as an additional fixed effect. The red dashed lines indicate the genome-wide significance threshold with α = 0.05 and Bonferroni correction for multiple testing (p = 3.8 × 10^–9). Due to computational limitations, variants with a p-value > 0.05 were excluded from the plots

Fig. 3

Genetic correlations of 61 conformation traits with withers height in horses. a includes all investigated traits. b–d only include the traits for which a genome-wide significant association signal on Equus caballus chromosome 3 (ECA3) between 97 and 118 Mb was detected and show their correlations for all autosomes (b), only ECA3 (c) and all autosomes except ECA3 (d). Trait codes are described in Table 1

Fig. 4

Overlapping runs of homozygosity (ROH) at a quantitative trait locus (QTL) for withers height. Number of horses with a high (blue) and low (red) chromosomal genomically estimated breeding value (gEBV) for withers height for which an ROH was detected at the respective position within the QTL on horse chromosome 3