The frontline support team will be slow on the forum because we are occupied with the GATK Workshop on March 21st and 22nd 2019. We will be back and more available to answer questions on the forum on March 25th 2019.
ReadBacked Phasing to find real compound hets variants
Just to give some context: I have filtered my trio data with some scripting to only heterozygous (hets) variants that may constitute compound hets (i.e., if phase could be accurately inferred). This is essentially phasing the child data by transmission - for all the het variants seen in the child I looked at the father and mother vcfs and filtered relevant sites as follows:
- each het variant in child has to be in only and exactly one of the parents, so this excludes 1) hets present in both parents (these cannot be resolved) and 2) hets not present in any parent (not interested on those as I only want to analyse compound hets);
- selected genes with at least two of the above vars;
- selected genes with at least one het transmitted from the paternal side and one het from the maternal side.
My question is: can I use this filtered child vcf as my input for ReadBackedPhasing? For each of my genes that feature in the child vcf after the above filtering, I want to determine whether the variants seen within the gene are in the same haplotype or not. I am just not sure if I can do the phasing at this stage - is this alright? If I had to do the phasing early on with the raw vcf, I am not sure how would I maintain the correct phasing information when applying this filtering downstream to the phased vcf (i.e., as the phasing of a het variant is relevant to the previous PASS-ing het variant in the vcf?).
Help would be appreciated!
Thanks a lot,