Hi,
Has anyone tried filtering structural variants (SV) based on population specific allele frequency (AF) values (for example gnomAD-SV or phase 3 1000 genome SV)?
I have a set of SVs that I detected using a multipronged approach. For prioritising variants, I wanted to remove SVs that are common variants in my population of interest (AF>0.05). In order to get population specific AFs for my detected SVs, I came across gnomAD-SV here: https://www.ncbi.nlm.nih.gov/dbvar/studies/nstd166/
I downloaded the GRCh38 version of the dataset from this link: https://ftp.ncbi.nlm.nih.gov/pub/dbVar/data/Homo_sapiens/by_study/vcf/nstd166.GRCh38.variant_call.vcf.gz. I then tried to annotate my VCF with it using the VEP plugin StructuralVariantOverlap,however, I did not get any AF values (no errors though). I even tried using SnpSift annotate, but that did not work either.
Am I missing something here that I am not getting any variant overlap between the query and target VCFs? The genome versions used seem to be the same. Anyone who has successfully done the above work, or tried doing it, can they please let me know if I am doing it right, or I have to pre-process the dbVar VCFs in any way?
Hi,
You can try AnnotSV. It is a program designed for annotating and ranking Structural Variations.
Thanks for this. I am yet to try it. Not sure if the issue is something to do with that I detected by SVs using long reads and the other datasets are using short reads. Technically, that shouldn't be an issue as they should in theory detect almost the same SVs.