Isolation-by-distance and population-size history inferences from the coho salmon (Oncorhynchus kisutch) genome

Abstract

Coho salmon (Oncorhynchus kisutch) are a culturally and economically important species that return from multiyear ocean migrations to spawn in rivers that flow to the Northern Pacific Ocean. Southern stocks of coho salmon have significantly declined over the past quarter century, and unfortunately, conservation efforts have not reversed this trend. To assist in stock management and conservation efforts, we generated two chromosome-level genome assemblies and sequenced 24 RNA-seq libraries to better annotate the coho salmon genome assemblies. We also resequenced the genomes of 83 coho salmon across their North American range to identify nucleotide variants, characterize the broad effects of isolation-by-distance using a genome-wide association analysis approach, and understand the demographic histories of these salmon by modeling population size from genome-wide data. We observed that more than 13% of all SNPs were associated with latitude (before multiple test correction), likely an affect of isolation-by-distance. From demographic history modeling, we estimated that the SNP latitudinal gradient likely developed as recently as 8,000 years ago. In addition, we identified four genes each harboring multiple SNPs associated with latitude; all of these SNPs were also predicted to modify the function of the gene. Three of these genes have roles in cell junction maintenance and may be involved in osmoregulation. This signifies that ocean salinity may have been a factor influencing coho salmon recolonization after the last glaciation period generating the current pattern of variation in these three genes.

Allele surfing causes maladaptation in a Pacific salmon of conservation concern

“…We then investigated historical change in effective population size using the Sequentially Markovian Coalescent in SMC++ [44] based on the 71 whole genome sequences ( Fig S2D) and both a median mutation rate of 0.8e −9 mutations/bp/generation and a mean of 1.25e −8 mutations/bp/generation as estimated for a closely related salmonid; the Atlantic salmon (J. Wang, personal communication). This analysis revealed: i) an expansion of most populations approximately 12-20 KyA, concomitant with postglacial recolonization, ii) a slow and steady decline in the Thompson R. ( Fig S2D) ), and iii) a split time between all pairwise combinations of populations (median = 16,3 KyA, range = 6,7KyA - 61KyA, Fig S3 ) compatible with the onset of postglacial population expansion following glacial retreat [38]. Using the mean mutation rate yielded more recent estimates of split times (median = 9,6 KyA, Fig S3 ) (min = 5 KyA – max = 18 KyA).…”

“…Previous studies showed that this population is highly divergent from any others [35,36,52] and displayed an increased rate of long runs of homozygosity [38]. It has been subjected to extensive hatchery enhancement (from a single population) to circumvent the decline of Coho salmon in the Thompson drainage [53].…”

“…We further demonstrated that one population, the Salmon River (Thompson R. watershed) displayed an increased fixed load and decreased selection efficacy compared to southern populations. Previous studies showed that this population is highly divergent from any others [35,36,52] and displayed an increased rate of long runs of homozygosity [38]. It has been subjected to extensive hatchery enhancement (from a single population) to circumvent the decline of Coho salmon in the Thompson drainage [53].…”

“…We used the TEs annotation file from repeatmasker [89] (made available on NCBI for the reference genome [38]) and tested for difference in the length of TEs between diploid and tetraploid regions, after correcting for the difference in chromosome length.…”

“…A salient feature of all salmonids is that their ancestor have undergone a whole genome duplication ~80-100MyA [37] and is still on its path to rediploidization. As a consequence, approximately 8% of the Coho genome displays residual tetraploidy [38]. Theoretically, these regions should display an increased effective population size (4 N e instead of 2 N e ) and therefore a higher efficacy of selection.…”