awk gff column change

Question

[rob234king]

I have had to split by fasta genome into chunks to speed up annotation using maker as we did not have mpi up and chromosomes were very large. I need to add the last value in the string of the first column to columns 4 and 5 to correct the starting and finish positions. I need to do this to multiple files so ideally parse the value rather than manually enter the value but even that would be helpful. I think this should be quite easy awk but still not familiar with it, any ideas?

Examples line:
Chr13_pilon_pilon_8000000       maker   gene    1257    1527    .       +       .       ID=maker-Chr13_pilon_pilon_8000000-exonerate_est2genome-gene-0.0;Name=evgtrinLocGG_7725c278g1t1-gene

Examples output:
Chr13_pilon_pilon_8000000       maker   gene    8001257    8001527    .       +       .       ID=maker-Chr13_pilon_pilon_8000000-exonerate_est2genome-gene-0.0;Name=evgtrinLocGG_7725c278g1t1-gene

Answer 1

[Jorge Amigo]

if the value you need to add to columns 4 and 5 is always behind a "_" character in the first column, then this perl code should work:

perl -lane 'if (/^\S+_(\d+)/) { $F[3] += $1; $F[4] += $1; print join "\t", @F}' input.txt

Answer 2

[Jianyu]

awk '{n=split($1,a,"_"); print $1,$2,$3,a[n]+$4, a[n]+$5,$6,$7,$8,$9}' input.txt

[Jorge Amigo]

not caring about the number of total columns:

awk '{n=split($1,a,"_"); $4+=a[n]; $5+=a[n]; print}' input.txt

not caring about the number of total columns, plus considering that both input and output are tabulated:

awk 'FS=OFS="\t"{n=split($1,a,"_"); $4+=a[n]; $5+=a[n]; print}' input.txt