

The GT-seq panel exhibited high self-assignment accuracy (93.3%) and perfect assignment of individuals not included in the baseline to their geographic basin, migratory form, and reproductive ecotype of origin. Here, we developed a multi-purpose Genotyping-in-Thousands by sequencing (GT-seq) panel of 288 targeted single nucleotide polymorphisms (SNPs) to enable accurate kokanee stock identification by geographic basin, migratory form, and reproductive ecotype across British Columbia, Canada. nerka in some systems across their pan-Pacific distribution. Such is the case for kokanee, the freshwater resident form of sockeye salmon ( Oncorhynchus nerka ), which exhibits various reproductive ecotypes (stream-, shore-, deep-spawning) that co-occur with each other and/or anadromous O. The ability to differentiate life history variants is vital for estimating fisheries management parameters, yet traditional survey methods can be inaccurate in mixed-stock fisheries. If successfully implemented, Nanopore sequencing will provide an alternative method to the large‐scale laboratory approach by providing mobile small batch genotyping to diverse stakeholders. Future improvements will focus on lowering turnaround time and cost, increasing accuracy and throughput, as well as augmentation of the existing baselines. However, poor representation of assayed salmon in the queried baseline data set contributed to poor assignment confidence on both platforms. Nanopore‐based SNP calls agreed with Ion Torrent‐based genotypes in 83.25%, but assignment of individuals to stock of origin only agreed in 61.5% of individuals, highlighting inherent challenges of Nanopore sequencing, such as resolution of homopolymer tracts and indels. Nanopore sequencing at‐sea yielded data sufficient for stock assignment for 50 out of 80 individuals. As long read sequencers are not optimized for short amplicons, we concatenate amplicons to increase coverage and throughput. Here we evaluate utility and comparability to established GSI platforms of at‐sea stock identification of coho salmon (Oncorhynchus kisutch) using targeted SNP amplicon sequencing on the minION platform during a high‐sea winter expedition to the Gulf of Alaska. However, recent advances in third‐generation single‐molecule sequencing platforms, such as the Oxford Nanopore minION, provide base calling on portable, pocket‐sized sequencers and promise real‐time, in‐field stock identification of variable batch sizes. The sequencing platforms currently applied require large batch sizes and multiday processing in specialized facilities to perform genotyping by the thousands. Genetic stock identification (GSI) from genotyping‐by‐sequencing of single nucleotide polymorphism (SNP) loci has become the gold standard for stock of origin identification in Pacific salmon. With 23 regions defined by the CWT program, and individuals displaying an assignment probability < 0.85 excluded from the analysis, average regional assignment accuracy of individuals via GSI was 98.4% over all 23 regions.


Inclusion of individuals requiring only a single parental genotype for identification resulted in assignments of 2,101 individuals, with an accuracy of 99.95% (2,000/2,001) to population and 100.0% to age. The overall accuracy of assignment for 1,939 Coho Salmon to the correct population was 100%, and to correct broodyear within population was also 100%. Coho Salmon from 15 populations were assigned via parentage analysis that required the genotypes of both parents. In total, 20,242 individuals from 117 populations were genotyped at 304 single nucleotide polymorphisms (SNPs) via direct sequencing of amplicons. Parentage-based tagging (PBT) and genetic stock identification (GSI) were used to identify individual Coho Salmon (Oncorhynchus kisutch) to specific populations and broodyears.
