Fig. 3: Genome-wide association with urbanisation.
From: Continent-wide genomic signatures of adaptation to urbanisation in a songbird across Europe
a Manhattan plot showing the genotype–urbanisation (PCurb) association estimated with LFMM across the genome. The red and blue dotted lines show the 0.1 and 1% FDR significance thresholds, respectively. Red dots highlight “core urbanisation SNPs” that were also identified as urbanisation associated using BayPass (Supplementary Fig. 6). The heatmap below the Manhattan plot highlights the SNP density across the genome. b Correlation between the number of urbanisation-associated SNPs (from LFMM) and the number of genes per chromosome (linear model: F1,30 = 284.60, P < 2.2 × 10−16). Shaded area denotes the 95% CI. c Main axis of variation in a PCAGEA based on “core urbanisation SNPs”. Grey circles show individuals, and yellow dots and lines show the mean ± SD for urban and rural populations, nurban = 96 and nrural = 96 independent birds. d Effect sizes (partial η2) for the effects of “habitat” (urban vs rural) and “habitat × locality” interaction on urban–rural allele frequency, using linear model are shown for all significant LFMM SNPs on the x-axis and y-axis, respectively. “Core urbanisation SNPs” are highlighted in yellow circles. SNPs that lie above the dashed line show strong consistent shifts in allele frequencies across localities, whereas SNPs below the line show more variable allele frequency shifts across localities. Source data are provided as a Source data file.
