[chr11:58991970-59056399][DEL][SVLEN=64429bp]

[jzook]

chr11 58991970 Minda_17 N <DEL> . PASS SVLEN=64429;SVTYPE=DEL;SUPP_VEC=PB_r_60,ONT_r_45,ILL_MantaDEL:2:57679:57680:0:0:0,PB_severus_DEL1144,PB_ID_48673_2,ONT_severus_DEL1245,ILL_gridss186fb_11621h,ILL_994122814:2,PB_Sniffles2.DEL.1F8MA,ONT_ID_38471_2,ONT_Sniffles2.DEL.1F9MA

https://v2.genomeribbon.com/?session=https://42basepairs.com/download/s3/giab-data/ribbon-json/ribbon-hg008-cov-bedpe.json&locus=chr11:58991970#ribbon

[aganezov]

Appears to be a true somatic deletion.

Coverage changes are consistent with a deletion signature:

• 200 -> 100 for in short read dataset (via splitthreader),
• 94 -> 50 (left side) and 58 -> 103 (right side) in PB
• 51 -> 27 (left side) and 24 -> 47 (right side) ONT

1. left-side breakpoint position is challenging and is either at chr11:58,991,970 or at chr11:58,991,975. PB has a boundary at both 58,991,970 and 58,991,975, with ~1/2 of the reads supporting the deletion going to 1 or another boundary value. ONT supports the 58,991,975 boundary. Element supports 58,991,971 boundary. Alignment suggest chr11:58,991,970. Careful examination of the genomic context suggest alignment artifact for both ONT and PB reads that support further-out boundary, as the overhang sequence is CATCA (while ref is CCTCA). For PB the left breakpoint overhang reads support the the further-out right breakpoint, and non-overhang left reads support nearer right breakpoints, suggesting the deletion of 64429bp. ONT reads with left breakpoint overhang support near right breakpoint, suggesting 64424bp deletion. Assembly supports the 64429bp deletion (left version of left breakpoint and left version of right breakpoint). Element suggests 64425bp deletion, with left version (1+bp) of the left breakpoint and left version of the right breakpoint.

2. right-side breakpoint position appears to be at chr11: 59,056,399. As supported by ONT, element, and assembly alignments. PB has both 59,056,399 and 59,056,404 breakpoint alignments, thus supporting the mapping issues reason for imprecise right breakpoint position reporting.

Split-long read analysis suggests exact breakpoint resolution with Haplotype 1 being affected (single haplotype being affected is consistent with a somatic SV signature).

All HP 1 reads seems to be affected, suggesting a clonal deletion.

[jzook]

Since the assembly supports this call and the leftmost drop in coverage for HiFi supports both breakpoints, I think the breakpoints are exactly correct, and the difference in alignments appears to be homology in the breakpoints, though with one SNV difference in the left breakpoint.