Fig. 2

Co-association modules for Pinus contorta. a Correlations among environments measured by Spearman’s ⍴ plotted according to hierarchical clustering of environments. Abbreviations of the environmental variables can be found in Table 2. Note the general categories on the left side of the heatmap. b Hierarchical clustering of the absolute value of associations between allele frequencies (of SNPs in columns) and environments (in rows) measured by Spearman’s ⍴. c–f Each co-association network represents a distinct co-association module, with color schemes according to the four major groups in the data. Each node is a SNP and is labeled with a number according to its exome contig, and a color according to its module—with the exceptions that modules containing a single SNP all give the same color within a major group. Numbers next to each module indicate the number of distinct genes involved (with the exception of the Geography group, where only modules with five or more genes are labeled). g The pleiotropy barplot, where each bar corresponds to a gene, and the colors represent the proportion of SNPs in each co-association module. Note that gene IDs are ordered by their co-association module, and the color of contig-IDs along the x axis is determined by the co-association module that the majority of SNPs in that contig cluster with. Contigs previously identified as undergoing convergent evolution with spruce by Yeaman et al. [46] are indicated with an asterisk. Abbreviations: Temp, temperature; Precip, precipitation; freq, frequency