Fig. 2
Hot structurally variable sites enriched with cis-regulatory elements. a Distribution of structurally variable nucleotides between adjacent developmental stages. b Nucleotide composition of all structurally variable nucleotides. c The distribution of common (2,3,4,5-way) or specific (1-way) structurally variable regions in different comparisons. The structurally variable region that is specific to one comparison is termed as “1-way,” while 2-way means that the structurally variable regions is shared by two comparisons, 3-way means that it is shared by three comparisons, and so on. d The statistics of all structurally variable regions grouped by their commonality. e Ratio of observed counts and expected counts of all structurally variable windows in three segments: 5′ UTR, CDS, 3′ UTR. The ratio is calculated by observed counts divided by expected counts. Statistical significance of enrichment of structurally variable windows in 3′ UTR was carried out with Fisher’s exact; P value < 2.23 × 10− 308 for all comparisons. f Metagene profile depicts the sub-transcript distribution pattern of common structurally variable regions shared by at least two comparisons. g Scatter plot shows the significance and occurrences of RNA-binding motif enriched in common structurally variable regions at 3′ UTR shared by at least two comparisons. P values were calculated by Fisher’s exact test. Refer to method section “De novo motif discovery and enrichment analysis of structurally variable regions”