Fig. 4

RNA structurally variable elements in Elavl1a binding regions associate with maternal RNA stability. a Average icSHAPE reactivity change (4 h.p.f. and 6 h.p.f.) at the Elavl1a binding site of the two groups of transcripts. Group I: more structural, forming more structure at Elavl1a binding sites. Group II: less structural, forming less structure at Elavl1a’s binding sites. “More structural” group was defined as the average of icSHAPE reactivity of those Elavl1-binding sites at 6 h.p.f. is less than that at 4 h.p.f. and the differences were larger than 0.05 (lower icSHAPE reactivity indicates that RNA structure become less accessible to SHAPE reagent, thus become more compact structure). Statistical significance was calculated by paired t-test and set to be P < 0.05. While the “less structural” group is defined as the average of icSHAPE reactivity of those Elavl1-binding sites at 6 h.p.f., it was larger than that at 4 h.p.f. and the difference is also larger than 0.05, P < 0.05. b Cumulative distribution of the log₂ fold changes of the RNA level between two group transcripts with more or less structural Elavl1a binding sites during the period of 4 h.p.f. to 6 h.p.f.. P value was calculated using two-sided Wilcoxon test. c Pie chart depicting the proportion of maternal decay genes in transcripts with more structural Elavl1a binding sites, and the proportion of maternal stable genes in transcripts with less structural Elavl1a binding sites during the period of 4 h.p.f. to 6 h.p.f.. d, e IGV tracks and structure profile displaying icSHAPE (upper panels), iCLIP-seq (middle panel), and RNA-seq (bottom panel) read distributions in 3′ UTR of mylipa (d) and atp2b1a (e) mRNA. Binding motifs are indicated with red highlight