Fig. 4

Overview of the variant detection. A. The length distributions of insertions (top) and deletions (bottom). The maximum length was capped at 1600 bp. B. The flowchart shows the breakdown of the insertions. Dotted-line boxes represent the conditions of classification, and solid-line boxes are each class of insertions. C. The pie charts show the proportion of each class of insertions of NA18943 (left) and NA19240 (right). Insertions were decomposed into multiple classes: TRs (tandem repeats), TEs (transposable elements), TDs (tandem duplications), dispersed duplications, processed pseudogenes, NUMT, “deletions” in GRCh38, alternative sequences, satellites, and others. D. A processed pseudogene (RPLP0) detected in NA19240 is shown (in black). The picture was extracted from Genome Browser. The BLAT identity was 99.9% compared with the reference gene, and the alignment pattern of 5′ UTR showed a closer resemblance with the splicing variant shown at the bottom. E. An insertion detected in NA19240 matching the alternative sequence is shown (in black). The picture was extracted from Genome Browser. The breakpoint was 15 kbp upstream of HLA-DQB1. LoMA constructed part of the alternative sequence accurately (99.8% in BLAT identity)