Genomic predictions can accelerate selection for resistance against Piscirickettsia salmonis in Atlantic salmon (Salmo salar)

BMC Genomics

Research Article

Rama Bangera¹  René Figueroa¹  Jean P. Lhorente¹  José M. Yáñez²  Katharina Correa³ 

[1] Aquainnovo S.A, Talca 60, Puerto Montt, Chile;Aquainnovo S.A, Talca 60, Puerto Montt, Chile;Facultad de Ciencias Veterinarias y Pecuarias, Universidad de Chile, Santa Rosa 11735, La Pintana, Santiago, Chile;Facultad de Ciencias Veterinarias y Pecuarias, Universidad de Chile, Santa Rosa 11735, La Pintana, Santiago, Chile;

关键词: Genomic selection;    Salmon Rickettsial Syndrome;    Disease resistance;    Reliability;   

DOI  :  10.1186/s12864-017-3487-y

 received in 2016-08-12, accepted in 2017-01-11,  发布年份 2017

来源: Springer

【 摘 要 】

BackgroundSalmon Rickettsial Syndrome (SRS) caused by Piscirickettsia salmonis is a major disease affecting the Chilean salmon industry. Genomic selection (GS) is a method wherein genome-wide markers and phenotype information of full-sibs are used to predict genomic EBV (GEBV) of selection candidates and is expected to have increased accuracy and response to selection over traditional pedigree based Best Linear Unbiased Prediction (PBLUP). Widely used GS methods such as genomic BLUP (GBLUP), SNPBLUP, Bayes C and Bayesian Lasso may perform differently with respect to accuracy of GEBV prediction. Our aim was to compare the accuracy, in terms of reliability of genome-enabled prediction, from different GS methods with PBLUP for resistance to SRS in an Atlantic salmon breeding program. Number of days to death (DAYS), binary survival status (STATUS) phenotypes, and 50 K SNP array genotypes were obtained from 2601 smolts challenged with P. salmonis. The reliability of different GS methods at different SNP densities with and without pedigree were compared to PBLUP using a five-fold cross validation scheme.ResultsHeritability estimated from GS methods was significantly higher than PBLUP. Pearson’s correlation between predicted GEBV from PBLUP and GS models ranged from 0.79 to 0.91 and 0.79–0.95 for DAYS and STATUS, respectively. The relative increase in reliability from different GS methods for DAYS and STATUS with 50 K SNP ranged from 8 to 25% and 27–30%, respectively. All GS methods outperformed PBLUP at all marker densities. DAYS and STATUS showed superior reliability over PBLUP even at the lowest marker density of 3 K and 500 SNP, respectively. 20 K SNP showed close to maximal reliability for both traits with little improvement using higher densities.ConclusionsThese results indicate that genomic predictions can accelerate genetic progress for SRS resistance in Atlantic salmon and implementation of this approach will contribute to the control of SRS in Chile. We recommend GBLUP for routine GS evaluation because this method is computationally faster and the results are very similar with other GS methods. The use of lower density SNP or the combination of low density SNP and an imputation strategy may help to reduce genotyping costs without compromising gain in reliability.

【 授权许可】

CC BY   
© The Author(s). 2017

【 图 表 】

Fig. 5

Fig. 5

Fig. 1

Fig. 2

Fig. 4

Fig. 1

Fig. 2

12951_2016_246_Article_IEq1.gif

Fig. 1

Fig. 6

Fig. 7

12951_2015_155_Article_IEq73.gif

Fig. 6

12951_2015_155_Article_IEq72.gif

Fig. 3

Fig. 3

Fig. 1

Fig. 2

Fig. 2

Fig. 1

Fig. 2

Fig. 2

Fig. 4

Fig. 1

Fig. 3

Fig. 3

12947_2017_100_Article_IEq1.gif

Fig. 1

Fig. 1

Fig. 3

12951_2015_155_Article_IEq70.gif

Fig. 10

Fig. 2

Fig. 7

Fig. 5

Fig. 1

Fig. 19

Fig. 3

Fig. 6

Fig. 4

12951_2015_155_Article_IEq69.gif

Fig. 18

12888_2023_5206_Article_IEq3.gif

Fig. 17

12888_2023_5206_Article_IEq1.gif

Fig. 16

Fig. 1

Fig. 15

Fig. 3

Fig. 14

Fig. 5

Fig. 2

Fig. 13

Fig. 1

Fig. 1

Fig. 1

Fig. 12

Fig. 6

Fig. 3

Fig. 11

Fig. 8

Fig. 8

Fig. 1

Fig. 2