We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

1000Genome reference gvcf files for GenotypeGVCF

AswathyNAswathyN BangaloreMember

I am using GATK 2014.3-3.2.2-7-gf9cba99. I use GenotypeGVCF tool for joint genotying of my samples, where I consider 1000Genome reference gvcf files along with all the gvcf files of the batch. I observe that after this step when I split individual sample.vcf files there are more number of variants (1328 in my sample). But the same sample variants, after vqsr filteration, has 230 variants.

I ran the GenotypeGVCF step of the same batch of samples (gvcf files) without 1000Genome reference gvcf files. Then I got the same sample.vcf file with 419 variants after the spliting. The same file, after vqsr filteration, has 268 variants. My question is,

1)What is the impact of 1000Genome gvcf files in Joint genotyping?
2)Why the variant number is reduced after vqsr filteration, in the case where 1000Genome gvcf files were considered for joint genotyping?

Thanks in advance,



  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    It's difficult to say based on just this information. It sounds like the 1000 genome samples are causing you to call more potential variants, but after filtering the effect is minimal -- the results end up being fairly similar. That's actually a pretty good sign. You should check whether the calls that are different are marginal calls, and whether the technical profiles of the 1000 genome samples you used are similar to those of your own samples.

Sign In or Register to comment.