#!/usr/bin/perl -w
#
# illumina_split_read.pl
# A script to split a fastq file reads into two parts
#
# Martin.Trick@bbsrc.ac.uk, 21/1/2010

# Option to include a read number on command line if the required number of reads for the first part differs from the default of 40
# Script output two files,the program will exit if these files exist.
# the first file ends with .1 contains the first part of the reads, each read identifier ending with _1
# the second file ending with .2 contains the last part of the reads, each read identifier ending with _2


unless (@ARGV >= 1) {
  print "Usage: illumina_split_read.pl <fastq file> [split at base N (default 40)]\n";
  exit;
}

open (FQ, "<$ARGV[0]") or die "Couldn't open input fastq file ($!)\n";
my $out_1 = "$ARGV[0]" . ".1";
my $out_2 = "$ARGV[0]" . ".2";
if (-e $out_1 || -e $out_2) {
  print "Output files will be clobbered\n";
  exit;
}

open (OUT_1, ">>$out_1") or die "Couldn't open FH ($!)";
open (OUT_2, ">>$out_2") or die "Couldn't open FH ($!)";

my $split = $ARGV[1] || 40;

my $read_string;
my $quality_string;
my $read_name;

my $reads = 0;
my @record = ();

READ: while (<FQ>) {
  if ($.%4 == 1) {
    ($read_name = $_ ) =~ s/^\;//;
    chomp $read_name;
  }
  if ($.%4 == 2) {
    $read_string = $_;
  }
  next if ($.%4 == 3);
  if ($.%4 == 0) {
    $reads++;
    $quality_string = $_;
    print OUT_1 "$read_name" . "_1\n" . substr($read_string,0,$split) . "\n\+\n";
    print OUT_1 substr($quality_string,0,$split) . "\n";
    print OUT_2 "$read_name" . "_2\n" . substr($read_string,$split) . "+\n";
    print OUT_2 substr($quality_string,$split);
  }
}

print "Processed $reads reads\n";
close FQ;
close OUT_1;
close OUT_2;