Tabix ouputs a non-specified region VCF

Question

[Mehulsharma.253]

I ran the following command to extract a subset of regions in my VCF file using tabix:

 tabix -h myfile.vcf.gz "11:5247360-5247664" > myfilenew.vcf.gz

For some reason, no matter how I vary my regions, one particular site: 5247358 always comes up in my output. Why is this the case ?

P.S. noob in variant calling and analysing VCF files.

[ATpoint]

Can you show the full VCF line that comes up? Probably has to do with the length of the variant.

[finswimmer]

Hello and welcome Mehulsharma.253 ,

the quotation marks shouldn't be neccessary. Could you please show the variant(s) that you don't expect? I guess it will be an insdel that overlapt the region you specify.

fin swimmer

BTW: > myfilenew.vcf.gz will not create a compressed file. You have to pipe the output of tabix through bgzip:

$ tabix -h myfile.vcf.gz 11:5247360-5247664 | bgzip -c  > myfilenew.vcf.gz

Answer 1

[Pierre Lindenbaum]

one particular site: 5247358 always comes up in my output

isn't it a variant which is NOT a Single Nucleotide variant ? Is there a END attribute in the INFO column ?...

[Mehulsharma.253]

Just checked it. Yes, there is an END attribute. The END coordinate is overlapping my #POS range.

[Pierre Lindenbaum]

If an answer was helpful you should upvote it, if the answer resolved your question you should mark it as accepted.