The accuracy of genomic predictions for bacterial cold water disease resistance remains higher than the pedigree-based model one generation after model training in a commercial rainbow trout breeding population

Authors: Vallejo, Roger; CHENG, HAO - University Of California, Davis; FRAGOMENI, BRENO - University Of Connecticut; SILVA, RAFAEL M. - Zoetis; MARTIN, KYLE - Troutlodge, Inc; Evenhuis, Jason; Wiens, Gregory - Greg; Leeds, Timothy - Tim; Palti, Yniv

Submitted to: Aquaculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/8/2021
Publication Date: 12/15/2021
Citation: Vallejo, R.L., Cheng, H., Fragomeni, B.O., Silva, R.O., Martin, K.E., Evenhuis, J., Wiens, G.D., Leeds, T.D., Palti, Y. 2021. The accuracy of genomic predictions for bacterial cold water disease resistance remains higher than the pedigree-based model one generation after model training in a commercial rainbow trout breeding population. Aquaculture. 545:737164. https://doi.org/10.1016/j.aquaculture.2021.737164.
DOI: https://doi.org/10.1016/j.aquaculture.2021.737164

Interpretive Summary: Bacterial cold water disease causes significant mortality and economic losses in rainbow trout aquaculture.  Traditional selective breeding over several generations can be used to improve resistance to this disease in a population, and the amount of genetic improvement made each generation can be increased substantially by using state-of-the-art genomic selection approaches that improve the accuracy of genetic merit predictions.  However, the high costs and labor requirements for phenotyping and genotyping thousands of fish each generation present challenges for the practical implementation of genomic selection in commercial breeding programs.  This study assessed the impact of reducing the frequency of phenotyping and genotyping fish from a population on the accuracy of genetic merit predictions.  Scientists found that genetic merit predictions remained robust despite reducing the frequency to every other generation, and despite reducing the number of genetic markers genotyped per fish.  This study provides producers with a more practical, cost-effective approach for using genomic selection to expedite genetic improvement for resistance to bacterial cold water disease in their population of rainbow trout.

Technical Abstract: Bacterial cold water disease (BCWD) causes significant mortality and economic losses in salmonid aquaculture. Previously, we reported high genomic prediction (GP) accuracy of 0.72 for BCWD resistance in rainbow trout using the 57K Axiom SNP array. Due to the high cost of phenotyping and genotyping in rainbow trout breeding programs, it is paramount to know if acceptable GP accuracy can be obtained in the subsequent generation without retraining the GP model. In the current study, we found that the accuracy of GP without model retraining in the subsequent generation was reduced from 0.65 and 0.61 to 0.56 and 0.53 using a lower density array of 10K SNPs and a panel of 49 QTL-linked SNPs, respectively. Although markedly lower than the GP with model retraining, the GP accuracy without retraining was better than the pedigree-based model (PBLUP) with retraining (0.48) and substantially higher than the PBLUP model accuracy without retraining (0.22). Hence force, we conclude that genomic selection provides better predictions’ accuracy than the traditional PBLUP model even when ‘skipping’ one generation of model retraining. The weighted single-step GBLUP (wssGBLUP) and single-step Bayesian multiple regression BayesB (ssBMR-BayesB) had higher GP accuracy than single-step GBLUP (ssGBLUP), which is consistent with the oligogenic inheritance of BCWD resistance in this population. Imputation from the 10K array genotypes back to the 57K array genotypes did not improve the accuracy of GP, likely due to the high linkage disequilibrium in rainbow trout aquaculture breeding populations and the oligogenic architecture of BCWD resistance.