Very high proportion 0/1 genotype in DUP, BND and INV

[hui-liu]

Hi,

I called the SVs using ~15X ONT reads. I found that the DUP, BND, and INV contained a high proportion of 0/1 genotype, with very few 1/1. Do you think this is normal? The low depth of ONT reads would cause this problem?

zcat ${sample}.vcf.gz | grep -v "#" |grep -v IMPRECISE |  grep DUP | cut -f 10 | sed 's/:.*//' |sort | uniq -c
     53 ./.
   1493 0/0
   4140 0/1
    131 1/1

zcat ${sample}.vcf.gz | grep -v "#" |grep -v IMPRECISE |  grep BND | cut -f 10 | sed 's/:.*//' |sort | uniq -c
  21444 0/0
  54112 0/1
   8067 1/1

zcat ${sample}.vcf.gz | grep -v "#" |grep -v IMPRECISE |  grep INV | cut -f 10 | sed 's/:.*//' |sort | uniq -c
      5 ./.
    756 0/0
   2086 0/1
    216 1/1

zcat ${sample}.vcf.gz | grep -v "#" |grep -v IMPRECISE |  grep INS | cut -f 10 | sed 's/:.*//' |sort | uniq -c
  41400 0/0
 219823 0/1
 256862 1/1

zcat ${sample}.vcf.gz | grep -v "#" |grep -v IMPRECISE |  grep DEL | cut -f 10 | sed 's/:.*//' |sort | uniq -c
  69472 0/0
 296181 0/1
 396590 1/1

[fritzsedlazeck]

Can you give more information, whats the coverage here ? The low coverage would typically lead to more 1/1 mutations. However, I see why you are concerned. What version of Sniffles was that ?

[hui-liu]

The coverage is about 15x, and the version of Sniffles is 2.4.