Seascape genomics of eastern oyster (Crassostrea virginica) along the Atlantic coast of Canada

Abstract

Interactions between environmental factors and complex life-history characteristics of marine organisms produce the genetic diversity and structure observed within species. Our main goal was to test for genetic differentiation among eastern oyster populations from the coastal region of Canadian Maritimes against expected genetic homogeneity caused by historical events, taking into account spatial and environmental (temperature, salinity, turbidity) variation. This was achieved by genotyping 486 individuals originating from 13 locations using RADSeq. A total of 11,321 filtered SNPs were used in a combination of population genomics and environmental association analyses. We revealed significant neutral genetic differentiation (mean F _ST = 0.009) between sampling locations, and the occurrence of six major genetic clusters within the studied system. Redundancy analyses (RDAs) revealed that spatial and environmental variables explained 3.1% and 4.9% of the neutral genetic variation and 38.6% and 12.2% of the putatively adaptive genetic variation, respectively. These results indicate that these environmental factors play a role in the distribution of both neutral and putatively adaptive genetic diversity in the system. Moreover, polygenic selection was suggested by genotype-environment association analysis and significant correlations between additive polygenic scores and temperature and salinity. We discuss our results in the context of their conservation and management implications for the eastern oyster.

Keywords: RADSeq; aquaculture; genomics; genotype by sequencing; oyster; population genetics.

Figure 1

Map of sampling sites within New Brunswick (NB), Nova Scotia (NS) and Prince Edward Island (PEI). BOU: Bouctouche, COC: Cocagne, CRB: Caribou, CRQ: Caraquet, INK: Inkerman, MAL: Malpeque, MIR: Miramichi, MIS: Miscou, RIC: Richibucto, SHD: Shediac, SHM: Shemogue, SSI: Saint‐Simon, TAB: Tabusintac

Figure 2

UPGMA dendrogram based on pairwise F _ST values among 13 eastern oyster sampling sites with bootstrap values based on 1,000 replicates, clustering of individuals identified by DAPC for K = 8 to 10 and heatmap based on pairwise F _ST values among the 13 sampling sites. All analyses are based on 8,246 neutral SNPs. Abbreviations of sampling sites are presented in Table 1

Figure 3

Assignment of individual oysters to their original (grey circles) or another sampling site (white circles). Numbers in circles represent the percentage of individuals from a sampling sites assigned to another. Circle diameter is proportional to percentage. Numbers in parenthesis represent the percentage of individuals from a “current population” tagged as migrants from a “inferred population”

Figure 4

Redundancy analysis (RDA) for analyses performed on the 8,246 neutral SNPs (a) and the six loci putatively under divergent selection (b). Arrows represent significant spatial (dbMEMs) and environmental variables on axes 1 and 2. Sampling sites are represented by black filled circles. PC factors are positioned according to the top and right axes. Abbreviations of sampling sites are presented in Table 1

Figure 5

Redundancy analysis (RDA) for polygenic adaptation analyses performed using the 11,321 SNPs on significant axes, (a) axes 1 and 2, (b) axes 3 and 4, and (c) axes 5 and 6. Arrows represent environmental variables (MSS: mean surface salinity, NW2‐6: number of weeks with temperature between 2 and 6°C, NW6‐10: number of weeks with temperature between 6 and 10°C, NW10‐14: number of weeks with temperature between 10 and 14°C, NW14‐18: number of weeks with temperature between 14 and 18°C, NW18: number of weeks with temperature above 18˚C, MST: mean surface temperature, MIST: minimum surface temperature, MAMT: maximum monthly turbidity, MIMT: minimum monthly turbidity). Large coloured circles and small grey circles represent sampling sites and SNPs, respectively. Abbreviations of sampling sites are presented in Table 1

Figure 6

Correlations between additive individual polygenic score based on (a) mean surface salinity and 17 SNPs, (b) number of weeks with temperature between 2 and 6°C and 10 SNPs, (c) number of weeks with temperature between 6 and 10°C and 30 SNPs, (d) number of weeks with temperature between 10 and 14°C and 60 SNPs, (e) number of weeks with temperature between 14 and 18˚C and 27 SNPs, (f) number of weeks with temperature above 18°C and 18 SNPs, (g) mean surface temperature and 47 SNPs, (h) minimum surface temperature and 38 SNPs. Only significant correlations are presented. Correlation coefficient (R ²) and p‐values of the linear (a, b) or quadratic (c, d, e, f, g, h) models are presented for each variable