jzook / testHG008curation

1 stars 0 forks source link

[chr11:33692823-35183689][DUP][SVLEN=1490866bp] #227

Open jzook opened 1 week ago

jzook commented 1 week ago

chr11 33692823 Minda_15 N <DUP> . PASS SVLEN=1490866;SVTYPE=DUP;SUPP_VEC=PB_severus_BND1090_2,PB_r_58,ONT_severus_DUP1176,ONT_r_42,ILL_MantaDUP:TANDEM:179586:0:1:0:0:0,PB_ID_48996_2,ONT_ID_38900_2,ILL_2882131873:2,ILL_gridss184bf_3824h,PB_Sniffles2.DUP.149MA,ONT_Sniffles2.DUP.143MA

https://v2.genomeribbon.com/?session=https://42basepairs.com/download/s3/giab-data/ribbon-json/ribbon-hg008-cov-bedpe.json&locus=chr11:33692823#ribbon

aganezov commented 1 day ago

Appears to be a duplication, with a tandem dup signature though variant is too large to have even the longest reads span from left flank, through both copies, onto the right flank of the region.

Coverage changes are consistent with a duplication:

  1. left-side breakpoint position appears to be exact at chr11: 33,692,823 (both ONT, PB, and Element have clear boundary. Also supported by the assembly mapping.

  2. right-side breakpoint position appears to be at chr11:35,183,689 as supported by ONT, PB, and Element, and assembly alignments.

Screenshot 2024-10-08 at 10 59 35

Internal structure of a duplicated segments appears relatively intact, at least based on a rough coverage analysis.

Screenshot 2024-10-08 at 11 00 31

Screenshot 2024-10-08 at 10 42 39 Screenshot 2024-10-08 at 10 43 39

Unclear if in all cells, but a evidence points into that direction (or at least largely clonal), as the overall number of reads on HP supporting the breakpoint and supporting ref structure (needed for tandem-dup-like signature) seems relatively even.