hom/het definition from Allelic Depth counts

Question

[cocchi.e89]

I am working on some VCF files, I was wondering (I could not find this answer searching on the web) if there is a rule or definition to consider hom based on AD. A lot of variants are defined hom if they have very few REF alleles, but in some cases I saw that, as example, a 92,342 (ref, alt) is considered 1/1 (hom), and also some 56,227or 7,31 etc

I imagine it depends on some kind of ratio amongst total read depth and altered depth, but I was wonder which threshold is normally used.

Thanks in advance for any help!

Answer 1

[Vitis]

I used a rule of 0.1/0.9 for homo and hets (alt/total counts between 0.1 and 0.9 are considered hets), specifically for amplicon genotyping results. This is an arbitrary threshold. I think your criteria need to be based on platform (amplicon? WGS? SeqCapture? mRNA-Seq?) and combining genotype information from a completely different platform such as genotyping chips.

[cocchi.e89]

Thanks for your reply! Data is from WGS