For this purpose, we selected all cases where editing level (ratio of edited to total reads) exceeded a threshold of 10% with >20 reads per library.
We observed also a significant decrease, from 20.7% to 15.4%, in the average editing level of Alu sequences in the spontaneous differentiation (n = 7 edited Alu regions, p<0.016 according to Wilcoxon signed-rank test; Figure 2A).
We define 'heavily edited proteins' as those for which the cumulative recoding level, that is the editing level summed over all recoding sites, exceeds unity.
The bars above the x-axis indicate upregulated editing level, and those below indicate downregulated editing level in leaf tissues as compared to floral tissues.
The differential editing level (>20%) for each edit between leaf and floral tissues is shown.
Two edits, petL-56 and rps12-221 are in high editing level (>83%) both in leaf and floral tissues.
Three edits psbN-29, psbN-30 and ccsA-336 are in medium editing level (45 54%) in leaf, but poor (8 9%) in flower.
In contrast, petG-56 and rps3-583 are in high editing level (64 72%) in flower, but relatively low (30 35%) in leaf.
In addition, ccsA-652 and ycf2-3868 showed poor partial editing (11 16%) in leaf and much less editing level (0 3%) in floral tissue.
We discovered two new edits (rpoB-55; rpoC1-1638) from rpo transcripts, although their editing level was poor (~10%) (Additional file 1: Tables S2, S4).
In addition, leaf-specific editing was found in the clpP intron (71,815) and flower-specific editing in the clpP intron (72,384), though the editing level is poor.
