Search

Chinook salmon (Oncorhynchus tshawytscha) are migratory fish that are highly valued for subsistence, sport, and commercial fisheries throughout their native range. Populations of Chinook salmon in western Alaska have exhibited long-term declines, leading to restrictions on harvests. Management priorities require greater resolution for genetic stock identification (GSI) than is available with current methods. We leveraged RADseq, TaqMan, and GT-seq data originating from multiple sources, collected through time, to develop a set of GT-seq panels containing 1092 single nucleotide polymorphisms (SNPs) that improved GSI resolution in western Alaska for at-sea and in-river sampling. We generated a dense linkage map to ensure that markers selected for panels spanned the entire genome. In addition, we identified multiple RADseq markers that were associated with sex; these aligned to a 5-centimorgan (cM) region on the sex chromosome. Finally, we developed a bioinformatic pipeline to streamline analysis of GT-seq data that is capable of genotyping microhaplotypes and paralogs, both of which can improve GSI resolution over traditional single-SNP data. Our panels and pipeline provide tools for management agencies to rapidly and easily analyze large-scale genotyping projects.

Pink salmon, the most abundant Pacific salmon, have an obligate 2-year life cycle that leads to reproductively isolated even- and odd-year lineages. Using new and existing data, we examined the genetic structure of both lineages across their distributional range by genotyping 16 681 single-nucleotide polymorphisms for 383 individuals originating from seven pairs of even- and odd-year populations. Distinct differences in standing pools of genetic variation were identified between the lineages; we observed higher levels of heterozygosity, allelic richness, and significantly more private alleles in the odd-year lineage. However, the patterns of population structure were concordant between lineages: the Asian and northern Alaska populations displayed little differentiation but differed significantly from populations in southcentral Alaska and the Pacific Northwest. Our population structure results, in the context of known paleoecological information, suggest that both lineages occupied a northern Beringial refugium as well as a Cascadian refugium in North America during the Last Glacial Maximum. These results highlight the influence of historical patterns of habitat availability on contemporary population structure and support the hypothesis of a pre-glacial origin of the lineages.

A common challenge for fisheries management is resolving the relative contribution of closely related populations where accuracy of genetic assignment may be limited. An overlooked method for increasing assignment accuracy is the use of multi-SNP (single nucleotide polymorphism) haplotypes rather than single-SNP genotypes. Haplotypes increase power for detecting population structure, and loci derived from next-generation sequencing methods often contain multiple SNPs. We evaluated the utility of multi-SNP haplotyping for mixture analysis in western Alaska Chinook salmon (Oncorhynchus tshawytscha). Multi-SNP haplotype data increased the accuracy of mixture analysis for closely related populations by up to seven percentage points relative to single-SNP genotype data for a set of 500 loci; 90% accuracy was achievable with as few as 150 loci with multi-SNP haplotypes but required at least 300 loci with single-SNP genotypes. Individual assignment to reporting groups showed an even greater increase in accuracy of up to 17 percentage points when multi-SNP haplotypes were used. Haplotyping multiple SNPs shows promise to improve the accuracy of assigning unknown fish to population of origin whenever haplotype data are available.