jzook / testHG008curation

This repository is primarily for the Genome in a Bottle Consortium to curate structural variants in HG008
https://www.nist.gov/programs-projects/cancer-genome-bottle
1 stars 0 forks source link

[chr6:72397103-72397193][DEL][SVLEN=90bp] #191

Open jzook opened 3 weeks ago

jzook commented 3 weeks ago

chr6 72397103 Minda_121 N <DEL> . PASS SVLEN=90;SVTYPE=DEL;SUPP_VEC=PB_d_385,ONT_d_168,PB_severus_DEL6208,ONT_severus_DEL7158,ONT_Sniffles2.DEL.26DM5

https://v2.genomeribbon.com/?session=https://42basepairs.com/download/s3/giab-data/ribbon-json/ribbon-hg008-cov-bedpe.json&locus=chr6:72397103#ribbon

aquaskyline commented 2 weeks ago

Subclonal 90 bp deletion in a long LOH with 4/41 Pacbio reads all on -strand but 0 ONT read support

Most of the chr6 is at ~100x (implying LOH). The 90bp deletion is in a LOH. A het C>A in normal and hom C>A in tumor at chr6:72,397,128 that the 90bp deletion covers also suggest it's a LOH region.

Screenshot 2024-10-05 at 15 32 33

Screenshot 2024-10-05 at 15 27 37 Screenshot 2024-10-05 at 15 27 58

Found 4/41 Pacbio reads (all on -strand) but 0/27 ONT read support. The breakpoints are exact. The right-end breakpoint is further supported by 1 short read with soft-clipping. DRAGEN called the variant with exact breakpoints. Likely a correct subclonal 90bp deletion but can use ONT read confirmation.

Screenshot 2024-10-05 at 23 06 20

jzook commented 2 weeks ago

Agree that seems likely true subclonal del. Added in standard ONT and 9/29 reads support the del, so maybe just got unlucky with UL-ONT not supporting it

Screenshot 2024-10-07 at 4 18 20 PM