How to merge, join, concatenate the first two columns in a text file separated by underscore? - bash

Question

I have a text: SG_gen.txt file with multiple columns looking like this:
snp_CHR POS HG00096 HG00097 HG00099 HG00100 HG00101 HG00102 HG00103
snp_3 47609552 0 1 1 1 1 0 1
snp_3 47614413 0 1 1 1 1 0 1
snp_3 47616151 0 1 1 1 1 0 1
snp_3 47616155 0 1 1 1 1 0 1
snp_3 47617504 0 1 1 1 1 0 1
snp_3 47617679 0 1 1 1 1 0 1
...
I would like to join the first two columns: snp_CHR and POS by "_" and rename it to ID so that column would look like this:
ID
snp_3_47609552
snp_3_47614413
snp_3_47616151
...
This new ID column would be the first column and I would keep all those other columns HG00096, HG00097...I would not keep the original snp_CHR and POS. How would I do this?
I tried using:
awk '{print $0, $1 "_" $NF}' SG_gen.txt > SG_gen1.txt
but this didn't gave me my desired result.

Answer 1

this should do:
awk '{$1=(NR==1?"ID":$1"_"$2); $2=""}1' file
there will be extra white space, which can be normalized afterwards if needed.

Answer 2

If you want to try Perl. Note that this retains the spaces between other columns as in your sample input.
$ cat anika.txt
snp_CHR POS HG00096 HG00097 HG00099 HG00100 HG00101 HG00102 HG00103
snp_3 47609552 0 1 1 1 1 0 1
snp_3 47614413 0 1 1 1 1 0 1
snp_3 47616151 0 1 1 1 1 0 1
snp_3 47616155 0 1 1 1 1 0 1
snp_3 47617504 0 1 1 1 1 0 1
snp_3 47617679 0 1 1 1 1 0 1
$ perl -pe 's/^\s*//g; s/\s/_/; s/^\S+\s+\S+/ID/ if $.==1' anika.txt
ID HG00096 HG00097 HG00099 HG00100 HG00101 HG00102 HG00103
snp_3_47609552 0 1 1 1 1 0 1
snp_3_47614413 0 1 1 1 1 0 1
snp_3_47616151 0 1 1 1 1 0 1
snp_3_47616155 0 1 1 1 1 0 1
snp_3_47617504 0 1 1 1 1 0 1
snp_3_47617679 0 1 1 1 1 0 1
$