2.8.2 Gene set enrichment analysis
Gene set enrichment analysis (GSEA) was run in clusterProfiler (v3.18.1) (Yu et al., 2012) for each tissue separately to determine if sets of genes from the same gene ontology (GO) term/functional category showed significant, concordant differences between current-day and elevated CO2 conditions. Unweighted GSEA was run using the DESeq2 log2 fold-change values of all genes and the annotated GO terms as the ‘gene sets’. A minimum and maximum gene set size of 15 and 500, respectively, was used. GSEA determines if genes from the same functional category are significantly more likely to occur at the top or bottom of the log2 fold-change list and therefore whether these functional categories are up- or down-regulated at elevated CO2, respectively. P-values were adjusted for multiple comparisons using the Benjamini-Hochberg method and a significance threshold of padj < 0.05 was used. The GSEA results were imported into Cytoscape (v3.8.2) (Shannon et al., 2003) where EnrichmentMap (v3.3.1) (Merico et al., 2010) was used to create a network to visualise the functional enrichment results. All significant functional categories were included in the network as a circular node. Functional categories with > 0.25 similarity were linked by edges. Similar functional categories were manually grouped into clusters and labelled.