Comparative Analysis of Genomic and Pedigree-Based Approaches for Genetic Evaluation of Morphological Traits in Pura Raza Española Horses
The single-step best linear unbiased predictor (ssGBLUP) has emerged as a reference method for genomic selection in recent years due to its advantages over traditional approaches. Although its application in horses remains limited, ssGBLUP has demonstrated the potential to improve the reliability of...
| Autores: | , , , , , , , , |
|---|---|
| Tipo de documento: | artigo |
| Estado: | Versão publicada |
| Data de publicação: | 2025 |
| País: | España |
| Recursos: | Universidad de Sevilla (US) |
| Repositório: | idUS. Depósito de Investigación de la Universidad de Sevilla |
| OAI Identifier: | oai:dnet:idus________::ad190c2254b33ae7834b4e6b043405b8 |
| Acesso em linha: | https://hdl.handle.net/11441/186122 https://doi.org/10.3390/genes16020131 |
| Access Level: | Acceso aberto |
| Palavra-chave: | SNP-genotyping Single-step GREML Estimated breeding value Reliability Horses |
| Resumo: | The single-step best linear unbiased predictor (ssGBLUP) has emerged as a reference method for genomic selection in recent years due to its advantages over traditional approaches. Although its application in horses remains limited, ssGBLUP has demonstrated the potential to improve the reliability of estimated breeding values in livestock species. This study aimed to assess the impact of incorporating genomic data using single-step restricted maximum likelihood (ssGREML) on reliability (R2) in the Pura Raza Española (PRE) horse breed, compared to traditional pedigree-based REML. Methods: The analysis involved 14 morphological traits from 7152 animals, including 2916 genotyped individuals. Genetic parameters were estimated using a multivariate model. Results: Results showed that heritability estimates were similar between the two approaches, ranging from 0.08 to 0.76. However, a significant increase in reliability (R2) was observed for ssGREML compared to REML across all morphological traits, with overall gains ranging from 1.56% to 13.30% depending on the trait evaluated. R2 ranged from 6.93% to 22.70% in genotyped animals, significantly lower in non-genotyped animals (0.82% to 12.37%). Interestingly, individuals with low R2 values in REML demonstrated the largest R2 gains in ssGREML. Additionally, this improvement was much greater (5.96% to 19.25%) when only considering stallions with less than 40 controlled foals. Conclusions: Hereby, we demonstrated that the application of genomic selection can contribute to improving the reliability of mating decisions in a large horse breeding program such as the PRE breed. |
|---|