Splitting VCF/BCF file into seperate gene files

Question

[Lynne-95]

I have a multi-sample bcf file which I would like to split into smaller files per gene so I can use this for some downstream eQTL analysis.

I've started a bash script which pipes bcftools query -f '%SAMPLE\t%POS\t%REF\t%ALT\t%GT\n' into an awk script where I subsequently re-code the genotype, and then in theory make a multi-sample text file per gene (+/- 1MB), by running this in a while loop with a GTF file.

However this is proving difficult! And I wasn't sure if I've missed a function or package somewhere that can do this for me.

Any help is appreciated - I also can easily convert this file to vcf or split by sample depending on the approach.

[barslmn]

What exactly proving difficult? Whats the results you're getting and what you like to get?

Also you can use code sample to format your code.

Answer 1

[Pierre Lindenbaum]

I have a multi-sample bcf file which I would like to split into smaller files per gene so I can use this for some downstream eQTL analysis.

awk -F '\t' '($3=="gene") {printf("%s\t%d\t%s\n",$1,int($4)-1,$5);}' in.gtf | while read R
do
      bcftools view -O z -o "${R//[:-]/_}.vcf.gz" input.vcf.gz "${R}"
done