Mixed-stock analyses using genetic markers have informed fisheries management in cases where strong genetic differentiation occurs among local spawning populations, yet many fisheries are supported by multiple spawning stocks that are weakly differentiated. Freshwater fisheries exemplify this problem, with many harvested populations supported by multiple stocks of young evolutionary age and that are isolated across small spatial scales. As a result, attempts to conduct genetic mixed-stock analyses of inland fisheries have often been unsuccessful. Advances in genomic sequencing now offer the ability to discriminate among populations with weak population structure, by providing the necessary resolution to conduct mixed-stock assignment among previously indistinguishable stocks. We demonstrate the use of genomic data to conduct a mixed-stock analysis of Lake Erie's commercial and recreational walleye (Sander vitreus) fisheries and estimate the relative harvest of weakly differentiated stocks. We used RAD-capture (Rapture) to sequence and genotype individuals at 12,081  loci that had been previously determined to be capable of discriminating between western and eastern basin stocks (mean pairwise F_ST = 0.001) with 95% reassignment accuracy. An outcome not possible in the past with microsatellite markers. Genetic assignment of 1,075 fish harvested from recreational and commercial fisheries in the eastern basin indicated that western basin stocks supported the majority of the harvest during peak harvest (July – September). Composition of harvest changed seasonally, with eastern basin fish comprising much of the early season harvest (May – June). Clear spatial structure in stock-specific harvest existed; more easterly sites contained more individuals of east basin origin than did westerly sites. Our study provides important stock contribution estimates for Lake Erie fishery management and demonstrates the power of genomic data to facilitate mixed-stock analysis in exploited fish populations with weak population structure or limited existing genetic resources.

All of these fish were collected by Lake Erie management agencies, includes the: Ohio Department of Natural Resources, New York Department of Environmental Conservation, and Ontario Ministry of Natural Resources. Fin clips were preserved in 95% ethanol until DNA extraction using Qiagen DNEasy 96 kits. Individuals for baseline analysis were sampled by by electrofishing or gillnets over known spawning locations during spawning seasons 2012 - 2017. Individuals of unkwon origin (labeled in VCF viles as "MS") were were sampled during the spring, summer, and fall of the 2016-2018 fishing seasons. For both commercial and recreational fishery samples, anglers and fishers reported the grid location of their harvest on a map provided by creel agents at boat ramps/docks. The and total length (nearest 1 mm) of all sampled individuals was also recorded.

Rapture libraties were prepared using SbfI enzyme and the standard library preparation approach outlined in Ali et al. (2016) and detailed in Ackiss et al. (2020). Samples were sequenced using paired-end 150 BP sequencing on an Illumina HiSeq4000 at NovoGene (Sacramento, CA). Loci were then assembled de novo in STACKS v. 2.0 (Catchen et al. 2011; 2013) and a catalog of all putative loci was created in cstacks from data of all 96 individuals. Samples were demultiplexed using process_radtags (-e sbfI -i gzfastq -c -q -r --filter_illumina –bestrad), assembled de novo in ustacks (--disable-gapped, -m3, -M 3, -H, --max_locus_stacks 4, --model_type bounded, --bound_high 0.05), matched in sstacks (--disable gapped), converted to bam files using tsv2bam, and genotyped in gstacks. Finally, genotypes were called for all single nucleotide polymorphisms (SNPs) with a minor allele count greater than 3 (--mac 3) genotyped in 80% of all individuals (-R 0.8).

To retrieve the exact loci included in the Rapture panel the STACKS 2 catalog files can be used to generate identical SNP IDs as original research (included). However, data can be treated as a standard RAD-sequencing project and analyzed de novo. Additional metadata can be found on Geome.