Demographic history shaped geographical patterns of deleterious mutation load in a broadly distributed Pacific Salmon

Abstract

Understanding the impacts of current human activities on within-species genetic variation requires a thorough description of the historical factors that have shaped the genomic and geographical distribution of nucleotide diversity. Past and current conditions influencing effective population size have important evolutionary implications for the efficacy of selection, increased accumulation of deleterious mutations, and loss of adaptive potential under the nearly neutral theory. Here, we gather extensive genome-wide data that represent the extant diversity of the Coho salmon (Oncoryhnchus kisutch) to address three issues. First, we demonstrate that a single glacial refugium is the source of the majority of present-day genetic diversity, with minor but detectable inputs from secondary micro-refugia. We propose a scenario whereby several ancestral populations located south of the ice sheets expanded in postglacial time, swamping out most of the diversity from other putative micro-refugia. Following this expansion, we identify particular populations having undergone continuous declines in population size (Ne). Second, we combine multiple evidence from demographic modelling, analysis of recombination landscape, and genome-wide landscape of diversity to demonstrate that selection at linked sites and Hill-Robertson interference played a major role in shaping genetic diversity across the Coho salmon genome. Third, we demonstrate that this demographic history generated subtle differences in the load of deleterious mutations among populations, a finding that mirrors recent results from human populations. Taken together, we found considerable support for the joint contributions of demographic history and linked selection in the load of deleterious mutations. We suggest that these inferences should be better integrated in conservation genetics of managed fish species which currently focuses largely on within-population adaptation. Author SummaryReconstruction of a species' past demographic history from genome-wide data allows understanding how historical factors interact with intrinsic genomic properties to shape the distribution of genetic diversity along its genome and its geographic range. Here, we combine genotyping-by-sequencing and whole genome sequence data with demographic modelling to address these issues in the Coho salmon, a Pacific salmon species with rapidly declining census size in some parts of its range, notably in the south. Our demographic reconstructions indicate a linear decrease in genetic diversity towards the north of the species range, supporting the hypothesis of a major southern refugia for the Coho salmon and a northern route of postglacial recolonization. Accordingly, the number of candidate deleterious homozygous derived mutations was higher in northern populations. Demographic modelling also suggested the existence of cryptic refugia that may have been missed with the use of simpler summary statistics. We further showed that the species' genome was shaped by linked selection and biased gene conv...

Heterogeneity in effective size across the genome: effects on the Inverse Instantaneous Coalescence Rate (IICR) and implications for demographic inference under linked selection

“…To integrate more complex demographic scenarios, several recent studies considered interesting approaches, by assuming demographic models including two classes of N e along the genome, one for neutral loci and one for loci under linked selection. The proportion of the two classes and the ratio of N e between them were estimated together with other parameters of the demographic model, using either ABC Bernatchez, 2018, Roux et al, 2016) or a modification (Rougemont et al, 2020, Rougeux et al, 2017 of the diffusion approach implemented in the software ∂a∂i (Gutenkunst et al, 2009). The models described in the present study propose another direction for the development of demographic inference methods by accounting for linked selection through variable classes of N e along the genome, and using the IICR as summary statistic.…”

Demographic history and genomics of local adaptation in blue tit populations

“…2019) and hence enable their persistence and rapid redeployment in heterogeneous environments despite high gene flow (Lowry & Willis 2010; Sinclair-Waters et al 2017; Todesco et al 2019). In addition, scans for adaptive variants need to be interpreted in the light of variation of recombination along the genome (Stapley et al 2017) since it shapes the local extent of diversity and of differentiation (Cutter & Payseur 2013; Tine et al 2014; Burri et al 2015; Rougemont et al 2019). Lastly, genome scans results need to be interpreted in the light of the demographic history of populations.…”

Understanding population structure in an evolutionary context: population-specificFSTand pairwiseFST

“…Additionally, FST values between pairs of population samples (pairwise FST, Reynolds et al 1983) are routinely used to estimate population structure in human genetics (Pérez-Lezaun et al 1997;Ramachandran et al, 2005), conservation biology (Palsbøll et al 2007;Schwartz et al 2007), and evolutionary biology and ecology (e.g., Hemmer-Hansen et al 2013a;Therkildsen et al 2013a;Geraldes et al 2014;McKown et al 2014a;Jorde et al 2015;Rougemont et al 2020). Divergent selection in an environmental gradient can influence population structure (Nosil et al 2009;Orsini et al 2013), and researchers have examined geographic distance and habitat differences between populations as explanatory variables that impact population structure estimated based on genome-wide (average over loci) pairwise FST values (e.g., Rousset 1997;Bradbury and Bentzen 2007;Petrou et al 2014;Jorde et al 2015;Kitada et al 2017).…”

“…However, genome-wide population-specific FST is new among biologists. Despite the expected usefulness of genome-wide population-specific FST in evolutionary biology (Weir and Goudet 2017), applications have been sparse to date (e.g., Nicholson et al 2002;Weir et al 2005;Foll and Gaggiotti 2006;Buckleton et al 2016;Rougemont et al 2020).…”