Perl script for combining 2 files with multiple entries

Question

I have a tab-delimited text file like this:

contig11 GO:100 other columns of data
contig11 GO:289 other columns of data
contig11 GO:113 other columns of data
contig22 GO:388 other columns of data
contig22 GO:101 other columns of data

And another like this:

contig11 3 N
contig11 1 Y
contig22 1 Y
contig22 2 N

I need to combine them so that each 'multiple' entry of one of the files is duplicated and populated with its data in the other, so that I get:

contig11 3 N GO:100 other columns of data
contig11 3 N GO:289 other columns of data
contig11 3 N GO:113 other columns of data
contig11 1 Y GO:100 other columns of data
contig11 1 Y GO:289 other columns of data
contig11 1 Y GO:113 other columns of data
contig22 1 Y GO:388 other columns of data
contig22 1 Y GO:101 other columns of data
contig22 2 N GO:388 other columns of data
contig22 2 N GO:101 other columns of data

I have little scripting experience, but have done this where e.g. "contig11" occurs only once in one of the files, with hashes/keys. But I can't even begin to get my head around to do this! Really appreciate some help or hints as to how to tackle this problem.

EDIT So I have tried ikegami's suggestion (see answers) with this: However, this has produced the output I needed except the GO:100 column onwards ($rest in script???) - any ideas what I'm doing wrong?

#!/usr/bin/env/perl

use warnings;

open (GOTERMS, "$ARGV[0]") or die "Error opening the input file with GO terms";
open (SNPS, "$ARGV[1]") or die "Error opening the input file with SNPs";

my %goterm;

while (<GOTERMS>)
{
    my($id, $rest) = /^(\S++)(,*)/s;
    push @{$goterm{$id}}, $rest;
}

while (my $row2 = <SNPS>)
{
    chomp($row2);
    my ($id) = $row2 =~ /^(\S+)/;
    for my $rest (@{ $goterm{$id} })
    {
        print("$row2$rest\n");
    }
}

close GOTERMS;
close SNPS;

Answer 1

Look at your output. It's clearly produced by

• for each row of the second file,
  • for each row of the first file with the same id,
    • print out the combined rows

So the question is: How does you find the rows of the first file with the same id as a row of the second file?

The answer is: You store the rows of the first file in a hash indexed by the row's id.

my %file1;
while (<$file1_fh>) {
   my ($id, $rest) = /^(\S++)(.*)/s;
   push @{ $file1{$id} }, $rest;
}

So the earlier pseudo code resolves to

while (my $row2 = <$file2_fh>) {
   chomp($row2);
   my ($id) = $row2 =~ /^(\S+)/;
   for my $rest (@{ $file1{$id} }) {
      print("$row2$rest");
   }
}

---

#!/usr/bin/env perl

use strict;   
use warnings;

open(my $GOTERMS, $ARGV[0])
     or die("Error opening GO terms file \"$ARGV[0]\": $!\n");
open(my $SNPS, $ARGV[1])
     or die("Error opening SNP file \"$ARGV[1]\": $!\n");

my %goterm;
while (<$GOTERMS>) {
    my ($id, $rest) = /^(\S++)(.*)/s;
    push @{ $goterm{$id} }, $rest;
}

while (my $row2 = <$SNPS>) {
    chomp($row2);
    my ($id) = $row2 =~ /^(\S+)/;
    for my $rest (@{ $goterm{$id} }) {
        print("$row2$rest");
    }
}