[chr4:14331587-14332165][INS][SVLEN=578bp]

[jzook]

chr4 14331587 Minda_103 N <INS> . PASS SVLEN=578;SVTYPE=INS;SUPP_VEC=ONT_severus_INS17703,ONT_Sniffles2.INS.DBM3,ONT_i_277,PB_severus_INS15603,PB_Sniffles2.INS.D5M3,PB_i_665

https://v2.genomeribbon.com/?session=https://42basepairs.com/download/s3/giab-data/ribbon-json/ribbon-hg008-cov-bedpe.json&locus=chr4:14331587#ribbon

[vkpat]

Looks like an insertion from chr4:14331587:14,331,601 that matches with Dragen call.. Breakpoints is not exact here..

[jzook]

Inside a LINE, and looks like a partial LINE insertion based on the BLAT of the insert, but might be an interesting one for @jo-mc to take a look at. I think the differences in length between methods have to do with different ways this complex variant could be represented, as well as noise in the poly-A tail.

[jo-mc]

Yes looks like a truncated Line1 insertion. There is also a small deletion AGATTTAAGTCTTT, likely as a result from the insertion. No target site duplication typical of line1, probably as a result of the deletion. This small deletion causing the breakpoints to vary for clipped reads, and obscuring likely exact breakpoints. This appears to also be loss of heterozygosity as both revio and illumina show insertion (and deletion) for every mapped read, and 50x coverage for pacbio, 99x for illumina? . dragen vcf : PR:SR 69,0:74,0 0,40:0,32 illumina, showing deletion and insertion in every read: (NOTE: using chm13v2 ref) PacBio revio