The current GATK version is 3.6-0
Examples: Monday, today, last week, Mar 26, 3/26/04

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Powered by Vanilla. Made with Bootstrap.

HaplotypeCaller and Coverage. I get ./. but I have coverage

NadiaNadia Member Posts: 13

Hi, I'm calling Variants with HaplotypeCaller in a population of 2 Parents and 7 F1-individuals. After read backed phasing I'm combining the vcf files of my genotypes with CombineVariants. In the outfile I very often find "./.". I thought this means there is no coverage at a certain position. But at many positions I do have good coverage. Why do I then get ./.? Moreover I used FastaAlternateReferenceMaker and created a new reference sequence including the variants from the parents. In that case, after I run HC and do the phasing and combine variants steps, I only get "./." at positions where there is really no coverage (as I can see in my mappings).

Best Answer


  • NadiaNadia Member Posts: 13

    oh yes of course! That makes sense! I will try both. Thanks a lot!

  • NadiaNadia Member Posts: 13

    when I run HC over all my sample together and then do Read-backed phasing, shall I merge all my mappings and do the phasing or rather not merge them? And if not, does it make a difference in which order I give them to the Read-backed phasing walker? Should it be the same order as the the order of the genotypes on the raw variants file (the HC output). In the past I have experienced that the phasing didn't work so well when I called all my samples together

  • pdexheimerpdexheimer Member, Dev Posts: 543 ✭✭✭✭

    I haven't done much with phasing, but IIRC it only works on trios - which would mean you'll have to run it once for each offspring. I think you'll then have to merge the resulting VCFs (does CombineVariants respect phase? I have no idea). Since you'd effectively be splitting the VCF by sample before phasing, the order that you run them shouldn't matter at all

Sign In or Register to comment.