VCFtools whole genome data zipped

[visse226]

Hi all!
New here, introduction: I investigate inbreeding in an endangered species where I have whole genome data of several individuals and variants stored in VCF files.

I want to keep the file zipped (.vcf.gz) because of used memory etc.
Let's say I want to filter and keep only the SNPs with '--remove-indels' and using gzvcf, how do I make sure the output is not .vcf but also still compressed .vcf.gz ? And in the mean time nothing is unzipped? Because I get the idea VCFtools unzips everything in the mean time.. If I would use a pipeline and say gzip -c > output_file.vcf.gz this will not work right, because that compresses the output again but I do not want it to be uncompressed in the first place.

Help?!

Thanks

[wdecoster]

You can write the output to stdout and pipe that output directly to gzip to compress the generated result.

[visse226]

Hi, Thanks for quick response!

So that means with the pipe I would be doing it correctly? Or redirection operator?
I am new to all this, so I am not aware what the difference between the two is in terms of storing the whole output first in memory, or doing it line-by-line/saving memory?

Thanks!

[dsenalik]

You use both, a pipe to avoid creating an intermediate file, and then redirect, e.g.

vcftools --gzvcf in.vcf.gz --somemagicparameters --stdout | bgzip > out.vcf.gz

and afterwords create an index if you need it

tabix -p vcf out.vcf.gz

[visse226]

Thank you so much! Very helpful