A PANEL OF SINGLE NUCLEOTIDE POLYMORPHISM FOR PARENTAGE ASSIGNMENT IN RAINBOW TROUT  

In salmonids aquaculture, family-based selective breeding programs rely on accurate pedigree information for estimating genetic merits. The pedigree information can be tracked through the use of Passive Integrated Transponder (PIT) tags. However, PIT tags can be expensive, and cannot be used with small fish. DNA markers offer an alternative approach for pedigree tracking and microsatellites have been previously used for parentage assignments in rainbow trout. Recently, single nucleotide polymorphism (SNPs) have been used for parentage assignment in diverse species because they are highly abundant and are amenable for high throughput genotyping. Although SNP panels were used for parentage analysis in hatchery populations of steelhead, the anadromous form of rainbow trout, SNP panels for parentage assignment in farmed rainbow trout have not been reported to date. The objectives of this study were to develop a SNP panel for parentage assignment in rainbow trout and to evaluate the accuracy of parentage assignment using fish with known pedigrees.

Two groups of SNPs were targeted to develop a panel of Fluidigm SNP Type assays for parentage assignment in this study. The majority of the SNPs were selected from the 57K SNPs on the genotyping array to cover all 29 chromosomes of rainbow trout. We refer to them as "genome SNPs" hereafter. The rest of SNPs used to develop the panel for parentage assignment are trait-associated SNPs identified previously, including SNPs associated with stress response, growth and resistance to bacterial cold water disease in rainbow trout. Among the 128 SNPs selected for assay development, 102 (79.7%) working assays were developed. A panel of 95 SNP assays including 68 genome SNPs and 27 trait-associated SNPs was used to genotype 499 fish with known pedigrees from Troutlodge, Inc. and from the USDA/ARS National Center for Cool and Cold Water Aquaculture (NCCCWA). Parentage assignments matched perfectly with the known pedigrees when all 95 SNPs were used. Near perfect parentage assignment was obtained with the 68 genome SNPs, and the accuracy of parentage assignment was over 98% with a subset of 48 genome SNPs. Therefore, these SNP assays are robust and enable accurate parentage analysis in both Troutlodge and NCCCWA populations. Since SNPs with higher minor allele frequencies in diverse populations were selected for assay development, the SNP panel developed in this study should also work well in other rainbow trout populations.