Comparison between haplotype‐based and individual snp‐based genomic predictions for beef fatty acid profile in Nelore cattle
Authors:
- Fabieli Loise Braga FeitosaFabieli Loise Braga Feitosa
- Sabrina Thaise AmorimSabrina Thaise Amorim
- Elisa PeripolliEPElisa Peripolli
- Adrielle Matias FerrinhoAFAdrielle Matias Ferrinho
- Flavio Schramm SchenkelFlavio Schramm Schenkel
- Luiz Fernando BritoLBLuiz Fernando Brito
- Rafael EspigolanRERafael Espigolan
- Lucia Galvão de AlbuquerqueLALucia Galvão de Albuquerque
- Fernando BaldiFBFernando Baldi
- 22 December 2019
- journal article
- research article
- Published by Wiley in Journal of Animal Breeding and Genetics
- Vol. 137 (5) , 468-476
Abstract
The aim of this study was to evaluate the genomic predictions using the single‐step genomic best linear unbiased predictor (ssGBLUP) method based on SNPs and haplotype markers associated with beef fatty acids (FAs) profile in Nelore cattle. The data set contained records from 963 Nelore bulls finished in feedlot (±90 days) and slaughtered with approximately 24 months of age. Meat samples from the Longissimus dorsi muscle were taken for FAs profile measurement. FAs were quantified by gas chromatography using a SP‐2560 capillary column. Animals were genotyped with the high‐density SNP panel (BovineHD BeadChip assay) containing 777,962 markers. SNPs with a minor allele frequency and a call rate lower than 0.05 and 0.90, respectively, monomorphic, located on sex chromosomes, and with unknown position were removed from the data set. After genomic quality control, a total of 469,981 SNPs and 892 samples were available for subsequent analyses. Missing genotypes were imputed and phased using the FImpute software. Haplotype blocks were defined based on linkage disequilibrium using the Haploview software. The model to estimate variance components and genetic parameters and to predict the genomic values included the random genetic additive effects, fixed effects of the contemporary group and the age at slaughter as a linear covariate. Accuracies using the haplotype‐based approach ranged from 0.07 to 0.31, and those SNP‐based ranged from 0.06 to 0.33. Regression coefficients ranged from 0.07 to 0.74 and from 0.08 to 1.45 using the haplotype‐ and SNP‐based approaches, respectively. Despite the low to moderate accuracies for the genomic values, it is possible to obtain genetic progress trough selection using genomic information based either on SNPs or haplotype markers. The SNP‐based approach allows less biased genomic evaluations, and it is more feasible when taking into account the computational and operational cost underlying the haplotypes inference.Keywords
Funding Information
- Fundação de Amparo à Pesquisa do Estado de São Paulo (2009/16118‐5, 2015/25304‐8, 2016/24085‐3)
This publication has 38 references indexed in Scilit:
- Genome-wide association and prediction of direct genomic breeding values for composition of fatty acids in Angus beef cattleaBMC Genomics, 2013
- Composite Effects of Polymorphisms near Multiple Regulatory Elements Create a Major-Effect QTLPLOS Genetics, 2011
- Hot topic: A unified approach to utilize phenotypic, full pedigree, and genomic information for genetic evaluation of Holstein final scoreJournal of Dairy Science, 2010
- Genomic prediction when some animals are not genotypedGenetics Selection Evolution, 2010
- A relationship matrix including full pedigree and genomic informationJournal of Dairy Science, 2009
- The importance of haplotype length and heritability using genomic selection in dairy cattleJournal of Animal Breeding and Genetics, 2009
- Efficient Methods to Compute Genomic PredictionsJournal of Dairy Science, 2008
- Linkage Disequilibrium and Persistence of Phase in Holstein–Friesian, Jersey and Angus CattleGenetics, 2008
- Accuracy of Genomic Selection Using Different Methods to Define HaplotypesGenetics, 2008
- Haploview: analysis and visualization of LD and haplotype mapsBioinformatics, 2004