Genome-wide association analyses using a Bayesian approach for litter size and piglet mortality in Danish Landrace and Yorkshire pigs

BMC Genomics

Research Article

Ole Fredslund Christensen¹  Mogens Sandø Lund¹  Xiangyu Guo¹  Luc Janss¹  Guosheng Su¹ 

[1] Center for Quantitative Genetics and Genomics, Department of Molecular Biology and Genetics, Aarhus University, DK-8830, Tjele, Denmark;

关键词: GWAS;    QTL region;    Litter size;    Piglet mortality;    Pig;   

DOI  :  10.1186/s12864-016-2806-z

 received in 2015-09-23, accepted in 2016-05-27,  发布年份 2016

来源: Springer

【 摘 要 】

BackgroundLitter size and piglet mortality are important traits in pig production. The study aimed to identify quantitative trait loci (QTL) for litter size and mortality traits, including total number of piglets born (TNB), litter size at day 5 (LS5) and mortality rate before day 5 (MORT) in Danish Landrace and Yorkshire pigs by genome-wide association studies (GWAS).MethodsThe phenotypic records and genotypes were available in 5,977 Landrace pigs and 6,000 Yorkshire pigs born from 1998 to 2014. A linear mixed model (LM) with a single SNP regression and a Bayesian mixture model (BM) including effects of all SNPs simultaneously were used for GWAS to detect significant QTL association. The response variable used in the GWAS was corrected phenotypic value which was obtained by adjusting original observations for non-genetic effects. For BM, the QTL region was determined by using a novel post-Gibbs analysis based on the posterior mixture probability.ResultsThe detected association patterns from LM and BM models were generally similar. However, BM gave more distinct detection signals than LM. The clearer peaks from BM indicated that the BM model has an advantage in respect of identifying and distinguishing regions of putative QTL. Using BM and QTL region analysis, for the three traits and two breeds a total of 15 QTL regions were identified on SSC1, 2, 3, 6, 7, 9, 13 and 14. Among these QTL regions, 6 regions located on SSC2, 3, 6, 7 and 13 were associated with more than one trait.ConclusionThis study detected QTL regions associated with litter size and piglet mortality traits in Danish pigs using a novel approach of post-Gibbs analysis based on posterior mixture probability. All of the detected QTL regions overlapped with regions previously reported for reproduction traits. The regions commonly detected in different traits and breeds could be resources for multi-trait and across-bred selection. The proposed novel QTL region analysis method would be a good alternative to detect and define QTL regions.

【 授权许可】

CC BY   
© The Author(s). 2016

【 图 表 】

12864_2017_3938_Article_IEq8.gif

12864_2017_3670_Article_IEq2.gif

12864_2017_3910_Article_IEq2.gif

12864_2017_4274_Article_IEq4.gif

12864_2016_2806_Article_IEq5.gif

12864_2017_3496_Article_IEq1.gif

12864_2017_3655_Article_IEq15.gif

12864_2015_2296_Article_IEq86.gif

12864_2017_3661_Article_IEq3.gif