Fig. 4: NaP-TRAP captures general and dynamic 5’ UTRs cis-regulation.
a The translation of the 5’ UTR library at 2 hpf and 6 hpf using NaP-TRAP (two-sided Pearson’s R; N = 8,529 reporters). Reporters were divided into four groups based on their translation rank at both time-points: repressed, active, active post-ZGA, and repressed post-ZGA (orange, blue, pink, and green). b–i Fold-change enrichment and depletion for all pentamers in each group from (a) relative to the reporter library (one-sided hypergeometric test to calculate significance; Bonferroni corrected p-value threshold, p < 5 * 10−6; repressed (b), active (d), active post-ZGA (f), and repressed post-ZGA (h)). Using hierarchical clustering sequence motifs were generated from enriched pentamers. The cumulative distribution of translation for reporters enriched (red, top 20%) and depleted (blue, bottom 20%) in a representative motif were plotted (middle 80% gray). The significance in the difference of the distributions of translation was determined using a two-sided Mann-Whitney U-test ((c), p < 10−103; (e), p < 10−171; (g), p < 10−73; (i), p < 10−71). j Schematic detailing the library of all possible tetramer repeats separated by dinucleotide spacers. k Translation measurements of the validation library at 2 and 6 hpf (two-sided Pearson’s R; N = 195 reporters). Tetrameric reporters were labeled based on whether their repeat was enriched in the reporter groups described above (a). l Heat map showing the significance of the overlap between eCLIP motifs for RBPs expressed in the early embryo (top 10) and the motifs enriched in each class in panel (a) (one-sided hypergeometric test). m Cumulative distribution plot comparing the translation (2hpf / 6 hpf) of validation reporters identified as active post ZGA (pink) and repressed post ZGA (green) (two-sided Mann-Whitney U-test: p < 10−5). n Using STREME, motifs were generated from the reporter groups described above (a). The information content of representative motifs was plotted: repressed (b), active (d), active post-ZGA (e), and repressed post-ZGA (f). Displayed E-values reflect the output of STREME. Tomtom was utilized to compare motifs to a database of human RBP motifs (E-value < 0.05).
