Genome-wide Association Studies and Genomic Prediction for Resilience Traits to Aleutian Disease in Mink

Abstract Aleutian disease (AD) is a significant health concern for mink, resulting in substantial economic losses for the mink industry. Although phenotypic selection of AD-resilient mink is practiced by some mink farms, the genetic architecture of AD resilience and the potentials of genetic or genomic selectionfor AD resilient traits have not been widely explored. Thus, the objectives of this study were to conduct genome-wide association studies (GWAS) to identify genomic regions and genes associated with AD resilience and investigate the efficiency of different genetic/genomic methods in predicting breeding values. The studied individuals (n = 1,411) had genotypes from the Axiom Affymetrix Mink 70K single nucleotide polymorphism (SNP) panel, with phenotypes including two immune response traits measured by antigen-based enzyme-linked immunosorbent assay (ELISA-G) and iodine agglutination test (IAT) and two feed-intake-related resilience traits measured by the daily variation in feed intake (Varf) and proportion of off-feed days (DOF). The methods of genetic and genomic predictions were traditional pedigree-based best linear unbiased prediction (BLUP), genomic BLUP (GBLUP), and single-step genomic BLUP (ssGBLUP). From the GWAS, seventeen, eight, and seven significant (q < 0.01) SNPs were associated with ELISA-G, IAT, and DOF, respectively, and 141, 44, and 42 genes were annotated using these significant. These GWAS analyses help provide a better understanding of the genetic architecture underlying AD resilience in mink and the potential for improving AD resilience using marker-assisted selection. In genetic and genomic prediction analyses, ssGBLUP had the highest prediction accuracy among all methods for IAT (0.53) and Varf (0.32). BLUP outperformed all genomic methods and produced the highest prediction accuracies for ELISA-G (0.71) and DOF (0.22). In the future, the availability of larger training populations, together with declining genotyping and sequencing costs and improved comprehension of genome and variant functional annotations, ssGBLUP is expected to emerge as the most effective approach for genomic prediction of AD resilience in mink.