It is difficult to give you a turn-key solution since we do not know the format of your input data, but perhaps you can get some inspiration from the following:

use Modern::Perl;

my @two_d_array;
my %dictionary;

sub add_data {
    chomp @_;
    my $type       = shift;
    my $datapoints = shift;
    state $sequence;
    push @two_d_array, [ split ',', $datapoints ];
    $dictionary{$type} = $sequence++;
}

while ( my $species = <DATA> ) {
    my $breaks = <DATA>;
    add_data( $species, $breaks );
}

say "$_:\t@{ @two_d_array[ $dictionary{$_} ] }" for sort keys %diction
+ary;

__DATA__
Chimp
71,22,15,10,51
Dog
91,82,28,11,91
Horse
11,72,37,58,20
Human
21,42,63,24,16
Monkey
81,32,53,54,42
Pig
10,13,99,12,25
Rat
9,17,87,33,11
[download]

The subroutine expects two parameters: the first is the name of the species and the second is the string of datapoints ("breaks" as you call them). It pushes the array_ref of the "breaks" on the @two_d_array and puts the name of the species as a key in the $dictionary hash with value the index of the corresponding "breaks" in the @two_d_array.

As you can see in the output routine (line 20), we walk the $dictionary hash and pull out the corresponding "breaks" from the @two_d_array.