how to use unix tools to convert VCF genotypes like '1|1' to this '2'

Question

[curious]

I want every row to be an id and every column a genotype for a different sample:

id1 0 2 1
id2 0 2 1

I know I can do this in python, but trying to get better with unix because the tools are usually faster

bcftools query -f '%ID[\t%GT]\n'  my_vcf.vcf |  awk -F "|" '{for(i=1; i<=NF; i++) { print $i+$i }}'

I think this is almost there by I almost there but obviously I'm off. Any hints are greatly appreciated thank you.

[Medhat]

How is your output look like? I think you should have two for loops, the first loops the samples, and for each sample loop the GT

[curious]

just this command bcftools query -f '%ID[\t%GT]\n' my_vcf.vcf gives me this:

id1 0|0 1|1 1|0
id2 0|0 1|1 1|0

I hope to get this with awk or similar:

id1 0 2 1
id2 0 2 1

right now i just get this

2
0
0

Answer 1

[Zhilong Jia]

cat 1.vcf

id1 0|0 1|1 1|0  
id2 0|0 1|1 1|0

sed -e 's/0|0/0/' -e 's/1|1/2/' -e 's/1|0/1/' -e 's/0|1/1/' 1.vcf

id1 0 2 1
id2 0 2 1