G'day BeckyLynn,

Welcome to the monastery.

I was unable to determine, from either your code or description, whether your input file contained one or many ">Sequence n" lines. The following code handles any number of such lines.

#!/usr/bin/env perl

use strict;
use warnings;
use autodie;

my ($in_file, $out_file) = qw{temp_in.txt temp_out.txt};

open my $in_fh, '<', $in_file;
open my $out_fh, '>', $out_file;

my @protein;

while (<$in_fh>) {
    chomp;

    if (/^>(.*)$/) {
        print_reverse_decoy(\@protein, $out_fh);
        print $out_fh ">Reverse Decoy $1\n";
    }
    else {
        push @protein, $_;
    }
}

print_reverse_decoy(\@protein, $out_fh);

sub print_reverse_decoy {
    my ($protein, $out_fh) = @_; 
    print $out_fh "$_\n" for map { scalar reverse split '' } reverse @
+$protein;
    @$protein = (); 
    return;
}
[download]

A test run with this input:

$ cat temp_in.txt
>Sequence 1
ABCDEFG
HIJKLMN
>Sequence 2
OPQRST
UVWXYZ
>Sequence N
1234567890
!@#$%^&*()
[download]

Produces this output:

$ cat temp_out.txt
>Reverse Decoy Sequence 1
NMLKJIH
GFEDCBA
>Reverse Decoy Sequence 2
ZYXWVU
TSRQPO
>Reverse Decoy Sequence N
)(*&^%$#@!
0987654321
[download]

Notes:

• Your shebang line looks wrong. I think you want '#!/usr/bin/perl', not '#! usr/bin/perl'.
• See perlrun for preference of warnings pragma over -w switch.
• Consider using the autodie pragma. It saves having to repeatedly type '... or die "Some custom error message: $!";' code. It also saves having to maintain the messages or check whether any have been omitted.
• Use the 3-argument form of open. There are a number of benefits from doing this, including being able to pass filehandles to a subroutine (as I've done in the code I've shown here).
• See how I've used chomp. Note that I've needed no other code to match or remove newlines with regular expressions.
• See perlre: Modifiers for usage of the 'g' modifier.