Spatial genomic structure along uplift vs. non-uplift habitats
We performed several approaches to document the extent of genetic differentiation among uplifted versus non-uplifted sites and to characterize uplift associated genetic structuring in the holdfast epifauna. We used principal component analysis (PCA) to explore SNP variation among individuals for each species separately using the R package adegenet v.2.1.1 (Jombart 2008). We applied the spare non-negative matrix factorization (sNMF) algorithm implemented in the R package LEA v.2.8.0 (Frichot et al. 2014; Frichot & François 2015) to infer individual ancestry coefficients and delineate putative genomic clusters. We tested for 1-15 ancestral populations (K ) with 50 replicates per each K value and chose the best Kby using a cross-entropy criterion. To evaluate the robustness of our results we ran the analysis using four values for the alpha regularization parameter (1, 10, 100 and 1000). Bar plots were visualized using the R package pophelper v.2.3.0 (Francis 2017). We inferred phylogenetic relationships among populations by constructing neighbor joining trees using SplitsTree v.4.16.1 (Huson & Bryant 2006).