Supporting information
Methods S1: Details of the pipeline and summaries used forde novo transcriptome assembly.
Methods S2: Pipeline details for SNP calling and filtering using ddRAD-seq genotyping.
Methods S3 : Details of steps and summary statistics for construction of the co-expression networks.
Figure S1: a) PCA of 76 environmental variables from climateWNA for the normal 1981-2010 for the 10 sampled populations and for the study year 2019 for the common gardens. (b) Heatmap representing the proportion of times drought or freeze related environmental variables measured at the gardens in 2019 were outside the range experienced by the assayed population across a 60-year period. Garden environment was declared as an outlier if it fell outside the 95 percent confidence interval of the climatic condition experienced by the populations for each year. This evaluation was done on a year-by-year basis. Warmer (red) colours indicate higher drought intensity and colder (blue) colours indicate higher freeze events experienced by the gardens relative to the assayed populations.
Figure S2: a) Geographical range map of P. strobiformisand P. flexilis along with the populations sampled for ddRAD-seq genotyping. b) STRUCTURE plot of genomic ancestry for the 30 maternal trees represented by the mixed-sicold sampled for RNA-seq andc) Distribution of FST values for the 30 maternal trees.
Fig S3: Visualization of Q ST enriched modules at the cold and warm gardens with circles representing the nodes (transcripts) and lines representing the edges (correlation between transcripts). Transcripts classified as Cold-condA or Warm-condA are highlighted in color while the rest are represented by black circles. The location of the transcripts in the module are representative of their edge weights (i.e intra-modular connectivity).
Fig S4: Relationship between mean transcript expression and co-expression connectivity levels for the a) Warm andb) Wold garden. Transcripts were divided into bins of mean expression by their position in the co-expression network (core vs. periphery) to highlight their level of connectivity.
Figure S5: A) Evaluation of scale free topology at various soft-thresholding powers and B) clustering dendrogram of transcripts using a soft-thresholding of 8 for the Cold garden and 9 for the Warm garden.
Table S1 : Annotations for all 47,651 transcripts generated through EnTAP.
Table S2: Summary of GO terms and the enrichment analyses for three Q ST categories (Cold-condA, Warm-condA, Ad-Pl) and for all modules across the two gardens.
Table S3: Summary statistics for modules identified across the gardens. Columns represent eigengene association with survival and ancestry, Q ST enrichment score, annotation of the transcript with the highest module membership, preservation scores and difference in mean Q ST between transcripts classified in the core and periphery of a module (QST periphery - QSTcore).