This is a test version of Biostars. For the public version, visit https://www.biostars.org.
split vcf by individual

Hi all!

I have a vcf file which contains 2 regions (in different chromosomes) for a group of individuals. I'm trying to split this vcf just by individual. The file IDlist.txt contains the IDs of all the individuals. The command I'm using is the following:

for i in $(cat IDlist.txt);
do i="${i%\\n}";
    vcf-subset -c $i file1.vcf > ${i}_file1.vcf
done

The command is able to split correctly the file by individuals, but the output contains just information about the first region (and not the second). Could anyone help me figuring out what's going on? Or suggest any other command that works better? Thank you very much in advance!! :) :)

vcftools vcf-subset snp vcf
• 4,094 views • link
updated by kitobo6683 • kitobo6683 • written by biosol biosol

2 answers

A combination of parallel and bcftools can do this:

$ cat IDlist.txt | parallel 'bcftools view -s {}_file1.vcf > {}_file1.vcf'

fin swimmer

• 1 views • link
written by finswimmer finswimmer

Hi, another way is to use bcftools query and GATK SelectVariants. Code is here:

cat split_vcf.sh

Script looks:

#!/bin/bash

genome=~/path/to/reference/genome.fa

for sample in `bcftools query -l INPUT.vcf`
do
  for i in *.vcf
  do
    gatk SelectVariants -R $genome -V $i -O ${sample}_${i}_test -sn $sample
  done
done

First for cycle is generate just list of Samples use in INPUT.vcf (here you need to put path to VCF). Second for cycle is take all VCF in current directory and split them according to first list ($sample variable).

You can run it as:

make script executable: chmod +x split_vcf.sh

and run it:

./split_vcf.sh
• 1 views • link
updated by Ram Ram • written by Paul Paul

Log in to answer this question.

More posts like this