Skip to main content
Fig. 3 | Genome Biology

Fig. 3

From: 2passtools: two-pass alignment using machine-learning-filtered splice junctions increases the accuracy of intron detection in long-read RNA sequencing

Fig. 3

Junction metrics can identify genuine splice junctions. a Outline of the two-pass method. b The JAD metric can discriminate between annotated and unannotated splice junctions in simulated nanopore DRS reads. Inverse cumulative density plot showing the distribution of per-splice junction maximum JAD values for annotated (blue) and unannotated (orange) splice junctions. c Flowchart visualization of the first decision tree model. Nodes (decisions) and leaves (outcomes) are colored based on the relative ratio of real and spurious splice junctions. d Confusion matrix showing the ratios of correct and incorrect predictions of the first decision tree model on splice junctions extracted from simulated Arabidopsis read alignments

Back to article page