Search code examples
bashtabular

How to remove text from a pattern in a certain column

I have a table file such as :

qseqid  sseqid  pident  length  mismatch    gapopen qstart  qend    sstart  send    evalue  bitscore    strand

scaffold    EOG091B09QV:/path/path/Z    xx  a   1:8830-20153    74.3    144 0   1
scaffold    EOG091B09QV:/path/path/A    x   a   1:8830-20153    100.0   93  0   0
scaffold    EOG091B09QV:/path/path/Q    x   a   1:8830-20153    41.3    189 49  3
scaffold    EOG091B09QV:/path/path/U    x   a   1:8830-20153    87.5    48  6   0
scaffold    EOG091B09QV:/path/path/K    x   a   1:8830-20153    100.0   60  0   0

And the idea is simply to remove in the column sseqid the text after : and get: 

qseqid  sseqid  pident  length  mismatch    gapopen qstart  qend    sstart  send    evalue  bitscore    strand

scaffold    EOG091B09QV xx  a   1:8830-20153    74.3    144 0   1
scaffold    EOG091B09QV x   a   1:8830-20153    100.0   93  0   0
scaffold    EOG091B09QV x   a   1:8830-20153    41.3    189 49  3
scaffold    EOG091B09QV x   a   1:8830-20153    87.5    48  6   0
scaffold    EOG091B09QV x   a   1:8830-20153    100.0   60  0   0

I know that cut -f 1 -d ":" matches_species_strand_H.m8 can work but not in column specific.

Solution

  • awk is a good choice to handle column based text:

    awk 'sub(/:.*/,"",$2)+7' file

    will do the job: "remove the :.* from the 2nd column.