Ewan Birney


Bio::Tools::SeqWords - Object holding n-mer statistics for one sequence


Take a sequence object from eg, an inputstream, and creates an object for the purposes of holding n-mer word statistics about that sequence. The sequence can be nucleic acid or protein, but the module is probably most relevant for DNA. The words are counted in a non-overlapping manner, ie. in the style of a codon table, but with any word length. For overlapping word counts, a sequence can be 'shifted' to remove the first character and then the count repeated. For counts on opposite strand (DNA/RNA), a reverse complement method should be performed, and then the count repeated.

Creating the SeqWords object, eg:

  my $inputstream = Bio::SeqIO->new( -file => "seqfile", 
                                     -format => 'Fasta');
  my $seqobj = $inputstream->next_seq();
  my $seq_word = Bio::Tools::SeqWords->new(-seq => $seqobj);


  my $seqobj = Bio::PrimarySeq->new(-seq=>'[cut and paste a sequence here]', 
                                    -alphabet => 'dna', 
                                    -id => 'test');
  my $seq_word  =  Bio::Tools::SeqWords->new(-seq => $seqobj);

obtain a hash of word counts, eg:

  my $hash_ref = $seq_stats->count_words($word_length);

display hash table, eg:

  my %hash = %$hash_ref;
  foreach my $key(sort keys %hash)
    print "\n$key\t$hash{$key}";


  my $hash_ref = Bio::SeqWords->count_words($seqobj,$word_length);


Bio:SeqWords is a featherweight object for the calculation of n-mer word occurrences in a single sequence. It is envisaged that the object will be useful for construction of scripts which use n-mer word tables as the raw material for statistical calculations; for instance, hexamer frequency for the calculation of coding protential, or the calculation of periodicity in repetitive DNA. Triplet frequency is already handled by Bio::SeqStats.pm (author: Peter Schattner). There are a few possible applications for protein, eg: hypothesised amino acid 7-mers in heat shock proteins, or proteins with multiple simple motifs. Sometimes these protein periodicities are best seen when the amino acid alphabet is truncated, eg Shulman alphabet. Since there are quite a few of these shortened alphabets, this module does not specify any particular alphabet.

See Synopsis above for object creation code.


Mailing Lists

User feedback is an integral part of the evolution of this and other Bioperl modules. Send your comments and suggestions preferably to one of the Bioperl mailing lists. Your participation is much appreciated.

  bioperl-l@bioperl.org                 - General discussion
  http://bio.perl.org/MailList.html     - About the mailing lists

Reporting Bugs

Report bugs to the Bioperl bug tracking system to help us keep track the bugs and their resolution. Bug reports can be submitted via email or the web:



Derek Gatherer, in the loosest sense of the word 'author'. The general shape of the module is lifted directly from Peter Schattner's SeqStats.pm module. The central subroutine to count the words is adapted from original code provided by Dave Shivak, in response to a query on the bioperl mailing list. At least 2 other people provided alternative means (equally good but not used in the end) of performing the same calculation. Thanks to all for your assistance.


The rest of the documentation details each of the object methods. Internal methods are usually preceded with a _


 Title   : count_words
 Usage   : $word_count = $seq_stats->count_words($word_length); 
 or      : $word_count = $seq_stats->Bio::SeqWords->($seqobj,$word_length);
 Function: Counts non-overlapping words within a string
         : any alphabet is used
 Example : a sequence ACCGTCCGT, counted at word length 4,
         : will give the hash
         : ACCG 1, TCCG 1
 Returns : Reference to a hash in which keys are words (any length) of the
         : alphabet used and values are number of occurrences of the word 
         : in the sequence.
 Args    : Word length as scalar and, reference to sequence object if
         : required

  Throws an exception word length is not a positive integer
  or if word length is longer than the sequence.