Fig. 2
From: Retained introns in long RNA-seq reads are not reliably detected in sample-matched short reads
a Distribution of persistence \(P_{i,t}\) and representative transcript examples for iPSC. The number of introns (y-axis) having a given persistence value (x-axis) is shown as a dark black line; note that a large number of introns with \(P_i=0\) are omitted from this analysis. Along the line, gray circles indicate the \(P_i\) value corresponding to each of nine introns from representative transcript examples (each transcript is labeled by Ensembl ID, e.g., ENST00000446856.5). Read-level data is shown for each transcript as a colored matrix, where each row is a single long read assigned to the transcript and each column represents a given intron, and color indicates whether an intron is retained (light green), spliced out (dark green), or lacking sequence coverage (white) in a given read. b Distributions of properties of persistent and called RIs. Each panel contains a series of boxplots depicting the distribution of intron length (top, log-scale), % of bases in the intron that are G or C (2nd row), relative position in transcript (3rd row), and % of intron bases with overlapping annotated exons (bottom) for HX1 (left) and iPSC (right). The distribution of each of these features is shown for long-read persistent introns (“PacBio”, gray) and RIs called by each of the eight short read tools: MAJIQ (light pink), rMATS (light green), SUPPA2 (light blue), IRFinder-S (red), superintronic (yellow), iREAD (green), KMA (blue), and IntEREst (purple)