Genomic architecture of adaptive radiation and hybridization in Alpine whitefish

Abstract

Adaptive radiations represent some of the most remarkable explosions of diversification across the tree of life. However, the constraints to rapid diversification and how they are sometimes overcome, particularly the relative roles of genetic architecture and hybridization, remain unclear. Here, we address these questions in the Alpine whitefish radiation, using a whole-genome dataset that includes multiple individuals of each of the 22 species belonging to six ecologically distinct ecomorph classes across several lake-systems. We reveal that repeated ecological and morphological diversification along a common environmental axis is associated with both genome-wide allele frequency shifts and a specific, larger effect, locus, associated with the gene edar. Additionally, we highlight the possible role of introgression between species from different lake-systems in facilitating the evolution and persistence of species with unique trait combinations and ecology. These results highlight the importance of both genome architecture and secondary contact with hybridization in fuelling adaptive radiation.

Fig. 1. Lake systems and species are genomically distinct across the Alpine whitefish radiation.

a A map of the Alpine whitefish species, assigned to ecomorphs, sampled from each of the pre-Alpine lake systems, including Constance (yellow), Zurich/Walen (purple), Lucerne (pink), Thun/Brienz (green) and Biel/Neuchâtel (blue). b A qualitative diagram showing the ecological characteristics of each whitefish ecomorph (represented by different symbols), including relative spawning depth (indicated by position in the figure), diet (indicated by letter), and relative gill-raker count (indicated by −/+ symbols); commissioned fish illustrations by Verena Kälin. c A genomic PCA of all 91 Alpine whitefish based on a linkage-disequilibrium filtered SNP-set of 1,133,255 (a subset of our full 14,313,952 SNP dataset) which separates out the Thun/Brienz system from all other lakes on PC1 and each of the other lake systems from one another on PC2. d A maximum likelihood RAxML phylogeny produced using a thinned subset of 1,692,559 SNPs from all 99 sequenced whitefish individuals (nodes have bootstrap support ≥ 95/100 unless highlighted with red triangles; outgroup samples with known ecomorph assignment are denoted with black symbols; for ease of viewing the most distantly related outgroup C. albula is pruned from this tree). e An admixture analysis highlights the lake-system based population structure within the Alpine whitefish radiation, and shows that sympatric whitefish species are each other’s closest relatives (to best observe within and between lake-system-level population structure, K = 7 is shown; see Supplementary Fig. 2 for the range of CV error associated with other values of K and Supplementary Fig. 3 for admixture proportions of individuals from K = 2 to K = 10).

Fig. 2. A combination of genome-wide allelic variation and major effect loci underpins parallel ecomorph differentiation in the Alpine whitefish radiation.

a CSS scan (calculated between 12 ‘Balchen’ individuals and 12 ‘Albeli’ individuals, with each ecomorph group comprising four distantly related species) highlights genome-wide parallel allele-frequency shifts between ‘Balchen’ and ‘Albeli’ ecomorphs across the four lakes. Outlier CSS windows are shown in black. b PC1 calculated using linkage filtered SNPs from across the 1659 CSS outlier windows for all whitefish individuals separates whitefish within lakes and lake systems (rows separated by dashed lines). Widespread and less-widespread ecomorphs within the same lake are separated along the same axis. c Whitefish standard length plotted against CSS PC1 for all lakes together (black line; R² = 0.498, P = 8.06 × 10⁻¹⁵) and for each lake separately. Significant lake-system-specific regressions are coloured by lake system and range in R² from to 0.322 in Lake Lucerne (P = 0.01405) to 0.6925 in the lake Walen/Zurich system (P = 1.84 × 10⁻⁵). d Gill-raker count plotted against CSS PC1 for all lakes together including (black line; R² = 0.3921, P = 4.1 × 10⁻¹¹) or excluding (grey line; R² = 0.5107, P = 7.63 × 10⁻¹⁵) the outlier species C. profundus, and for each lake system separately. Significant lake-system-specific regressions are coloured by lake and range in R² from 0.3871 in the Lake Thun–Brienz system (including the outlier C. profundus; P = 1.11 × 10⁻⁴; when excluding C. profundus R² = 0.6051, P = 4.22 × 10⁻⁷) to 0.8113 in lake Lucerne (P = 3.47 × 10⁻⁷). See Supplementary Table 2 for all details regarding lake-specific statistics. e Allele frequencies for the SNP significantly associated with gill-raker count variation where all 91 Alpine whitefish are grouped by ecomorph (black symbols) compared to ecomorph-averaged gill-raker counts (red symbols); commissioned fish illustrations by Verena Kälin. f GWAS results for the gill-raker count and g sex for all 9,120,498 polymorphic SNPs within the Alpine whitefish radiation across the 90 individuals with corresponding phenotypes.

Fig. 3. Excess allele-sharing is widespread between whitefish species both within and between lake systems.

F-branch (f_b(C)) statistics across our dataset highlight excess allele-sharing between tips in the tree (which represent species or individuals when species were not monophyletic; horizontally arranged at the top of the figure) and each other tip (solid line) and node (dotted line) in the phylogenetic tree (vertically arranged on the left of the figure), compared to its sister branch. The associated lake and ecomorph of each tree tip is indicated by the symbol and colour (as in Fig. 1a). The redness of each cell in the matrix indicates the degree of excess allele-sharing between each tree tip (C) and each tip or node (b) with significant instances of excess allele-sharing, where the Z-score was >4.41 (equivalent to the Bonferroni multiple-testing corrected P-value of 0.01), are highlighted with a dot. For clarity, when a species within a lake or lake system is supported as monophyletic we have collapsed all of its individuals into a single tree tip. Grey shading indicates tests which cannot be carried out due to the topology of the tree. F-branch statistics associated with species of the three focal ecomorphs are highlighted with boxes in the matrix, including the large-pelagic ecomorph of which we have three species from three lake systems (black), pelagic-profundal ecomorph as a single species from Lucerne (pink) and the benthic-profundal ecomorph as a single species from Thun (green).