G'day TJCooper,

I've provided an example script below which shows some techniques that you might use for this. I've assumed your IN2 file just contains headers (your description wasn't 100% clear on that). If you're going to be reusing IN2, you may want to consider running the first part of the script separately, and storing the @all_order data: a database, Storable, or other serialisation methods are all possible candidates (depends on your requirements and what you have available).

#!/usr/bin/env perl

use strict;
use warnings;
use autodie;

use Data::Dump;

my @all_order; 

{
    open my $in2_fh, '<', 'pm_1204589_IN2';

    while (<$in2_fh>) {
        chomp;
        push @all_order, $_;
    }
}

for my $range ([0..0], [0..2], [1..3], [3..3]) {
    my %wanted = map { $_ => 1 } @all_order[@$range];

    {
        open my $in1_fh, '<', 'pm_1204589_IN1';

        local $/ = "\n>";

        while (<$in1_fh>) {
            chomp;
            substr $_, 0, 1, '' if $. == 1;
            my ($head, $seq) = split;
            $wanted{$head} = $seq if $wanted{$head};
        }
    }

    print "Range: @$range\n";
    dd \%wanted;
}
[download]

With these dummy files (yes, I know, the sequences are totally bogus — just used for demonstration purposes):

$ cat pm_1204589_IN1
>Z
ABC
>Y
DEF
>X
GHI
>W
JKL
[download]

$ cat pm_1204589_IN2
W
X
Y
Z
[download]

I get this output:

Range: 0
{ W => "JKL" }
Range: 0 1 2
{ W => "JKL", X => "GHI", Y => "DEF" }
Range: 1 2 3
{ X => "GHI", Y => "DEF", Z => "ABC" }
Range: 3
{ Z => "ABC" }
[download]

See also: Data::Dump.