#!/usr/local/bin/perl

$\= "\n";
use Data::Dumper;

# this is the list from which we will draw the random base pair - add more if you like :)

@bp = qw(a t g c);

# explicitly initialize the list of sequences
@sequences = ();

for (1..1000) {
  # set the sequence to an empty string
  $seq = "";
  for (1..4) {
    # add a random base pair
    $seq = qq{$seq$bp[rand(@bp)]};
  }
  push @sequences, $seq;
}

# this will be the hash that will store the count of each sequence
%sequence_count = ();

# iterate through the list of sequences, and for each sequence increment
# its count, stored in the hash, by one
for $seq (@sequences) {
  $sequence_count{$seq} = $sequence_count{$seq} + 1;
}

for $seq (keys %sequence_count) {
  print qq{sequence $seq seen $sequence_count{$seq} times};
}

for $seq (keys %sequence_count) {
  if ($seq =~ /aaa|ccc|ggg|ttt/) {
    print qq(3-homo polymer sequence $seq seen $sequence_count{$seq} times);
  }
}

%bp_count = ();

# method 1 – iterate across sequences, split each sequence into list of characters
for $seq (@sequences) {
  for $bp (split("",$seq)) {
    $bp_count{$bp} = $bp_count{$bp} + 1;
  }
}

print Dumper(\%bp_count);

%bp_count = ();

for $seq (keys %sequence_count) {
  for $bp (split("",$seq)) {
    $bp_count{$bp} = $bp_count{$bp} + $sequence_count{$seq};
  }
}

print Dumper(\%bp_count);

$sum = 0;
for $count (values %sequence_count) {
  $sum = $sum + $count;
}
print "averge sequence count is ",$sum / keys %sequence_count;

for $seq (sort {$sequence_count{$b} <=> $sequence_count{$a}} keys %sequence_count) {
  print qq{sequence $seq seen $sequence_count{$seq} times};
}