The positions of two deletions are identical. Why weren’t they merged into a single entry?

Structural Variation
1

Hello. The position of DEL_CHR11_3DCD2FD1 and DEL_CHR11_F213CF06 are identical. Why weren’t they merged into a single entry? Can I merge them?

Position11:134738783-134739009

I found similar cases as follows:

 bcftools query -f '%CHROM-%POS-%ALT-%INFO/END-%CHR2-%INFO/POS2-%INFO/END2-%INFO/SVLEN\n' -i "SVTYPE = \"DEL\"" ../../original/gnomad.v4.1.sv.sites.vcf.gz | sort | uniq -d
chr11-134738783-<DEL>-134739009-chr11-.-.-226
chr11-68233234-<DEL>-68233798-chr11-.-.-564
chr1-201209715-<DEL>-201209798-chr1-.-.-83
chr20-50022383-<DEL>-50022483-chr20-.-.-100
chr21-27038559-<DEL>-27038636-chr21-.-.-77
chr21-45484269-<DEL>-45484471-chr21-.-.-202
chr3-177242504-<DEL>-177242908-chr3-.-.-404
chr3-49259693-<DEL>-49259766-chr3-.-.-73
chr5-110233-<DEL>-110378-chr5-.-.-145
chr5-124769362-<DEL>-124769439-chr5-.-.-77
chr6-140634236-<DEL>-140634421-chr6-.-.-185
chr6-87009825-<DEL>-87009902-chr6-.-.-77
chr7-75767243-<DEL>-75767347-chr7-.-.-104

Thank you for your help.

Nobutaka

2

Hi Nobutaka,

Thank you for pointing this out. We noticed this issue after the data release—the duplicated variants remain in the dataset because short-read sequencing often struggles to accurately detect structural variants in repetitive regions. For interpretation, I recommend focusing on variants with higher quality scores (the “QUAL” column in the VCF, or “Quality score” in the browser).

I hope this helps!

1 Like