vcftools --TajimaD does not start at 0 and misses variants

Question

[grey]

I have used vcftools --TajimaD to calculate Tajima's D statistics with other species easily in the past. However, I am running this now on a new vcf and getting strange results:

1. the output file doesn't start at 0, but rather at 539700. There is nothing special about this number. The first variant in the VCF file is at position 36 and so forth. See screenshot below
2. also, even for windows where it does make some calculations, it is missing variants that are clearly present in the VCF.

In case its helpful here is vcftools output

Indexing the vcf (tabix) did not solve.

Anyone encountered this before?

Answer 1

[grey]

SOLVED:

in the vcf-merge command used to create the vcf file I added --ref-for-missing 0/0

Perhaps vcftools --TajimaD doesn't like having missing values (ie ./.)?

[Ram]

Please accept your answer using the check mark on the left to mark the post as solved.

[grey]

it gives error "you can not accept your own post"

[Ram]

That's odd. I think it could be a bug with the new site. I'll accept your answer for you and alert the admin to this bug.