Fig. 2: Alternative splicing plays a smaller but complementary role to whole gene expression in seasonal plasticity.
From: Alternative splicing in seasonal plasticity and the potential for adaptation to environmental change
Results presented are for the abdomen, similar patterns were found in the thorax samples (Supplementary Fig. 2–6). a There was significantly more pairwise overlap (purple) of differentially spliced (DS, differential exon expression; red) and differentially expressed (DE, differential whole gene expression; blue) gene sets for the main effects of season, (one-sided Fisher’s exact test, OR = 1.45, P = 1.5 × 10−2, Supplementary Table 4) and family (one-sided Fisher’s exact test, OR = 1.57, P = 1.70 × 10−7) than predicted from the size of the filtered datasets, while genes with a season-by-family interaction (SxF) in splicing or expression did not overlap at all. b Effect sizes of differential splicing (e.g., relative exon log fold changes, red) were generally smaller than those of differential expression (e.g., whole gene log fold change, blue). For seasonal comparisons, fold change represents the change in expression from dry to wet. For family comparisons and SxF interaction, we used the maximum absolute fold change among all families as a proxy for fold change. c There was no relationship between whole gene log fold change and exon log fold changes when compared between season, among families or SxF. d Euler diagrams show that very few gene ontology (GO) terms enriched for genes that are differentially expressed (blue), both differentially expressed and differentially spliced (purple), or only differentially spliced (red) overlapped between these gene sets. Circle sizes correspond to the total number of enriched GO terms (two-sided Fisher’s Exact Tests, parentChild algorithm, p-value < 0.05), and in cases where the number of shared terms is very small, the number has been placed adjacent to the intersection. Source data are provided as a Source Data file.
