formatting VCF file for analyses

Question

[Matteo Ungaro]

Hi there,

I'm working with a VCF file of this type, see image below

However, I need a tool to handle this VCF so that I can genotype a particular human individual. The problem is that this tool doesn't handle the '.' for the individuals I have within my pangenome... I need those to be '.|.' instead.

I've tried this awk command

awk 'FNR > 719 {sub(/[[:space:]].[[:space:]]/, ".|."); print $1, $2, $3, $4, $5, $6, $7, $8, $9, $10, $11, $12, $13, $14, $15, $16, $17, $18, $19, $20}' pangenome_ref_guided_GRCh38.vcf > temp.vcf

but what it does is simply substituting the first '.' — green arrow, leaving the following ones untouched — red arrows. This behaviour is actually the opposite of what I need, as the '.' which need to be changed to a '.|.' are actually the ones in the section within red lines.

Thanks in advance, and sorry but I'm new to the use of awk and to handling VCF in general.

Answer 1

[Pierre Lindenbaum]

not tested

awk -F '\t' '/^#/ {print;next;} {OFS="\t";for(i=10;i<=NF;i++){ if($i==".") $i=".|."; } print;}' in.vcf

[Matteo Ungaro]

Hey @Pierre Lindenbaum,

Thanks a lot! It worked perfectly. If I may, could I ask you a questions about the command you wrote?

What the /^#/ does specifically, in my book it says it uses an "alternative form" for certain control letters; however, I'm not quite sure if that's the case here. Thanks again this approach also kept the first lines for the file, which I would have eventually need to add back.

[Pierre Lindenbaum]

 '/^#/     {                 // when a line starts with the regular expression 'start-with' hash
           print;            // print the whole line
           next;             // skip the other patterns, read the next line
            }