Ewan Birney


SimpleAlign - Multiple alignments held as a set of sequences


  # use Bio::AlignIO to read in the alignment
  $str = Bio::AlignIO->new('-file' => 't/data/testaln.pfam');
  $aln = $str->next_aln();

  # some descriptors
  print $aln->length, "\n";
  print $aln->no_residues, "\n";
  print $aln->is_flush, "\n";
  print $aln->no_sequences, "\n";
  print $aln->percentage_identity, "\n";
  print $aln->consensus_string(50), "\n";

  # find the position in the alignment for a sequence location
  $pos = $aln->column_from_residue_number('1433_LYCES', 14); # = 6;

  # extract sequences and check values for the alignment column $pos
  foreach $seq ($aln->each_seq) {
      $res = $seq->subseq($pos, $pos);
  foreach $res (keys %count) {
      printf "Res: %s  Count: %2d\n", $res, $count{$res};


SimpleAlign handles multiple alignments of sequences. It is very permissive of types (it won't insist on things being all same length etc): really it is a SequenceSet explicitly held in memory with a whole series of built in manipulations and especially file format systems for read/writing alignments.

SimpleAlign basically views an alignment as an immutable block of text. SimpleAlign *is not* the object to be using if you want to perform complex alignment manipulations.

However for lightweight display/formatting and minimal manipulation (e.g. removing all-gaps columns) - this is the one to use.

SimpleAlign uses a subclass of Bio::PrimarySeq class Bio::LocatableSeq to store its sequences. These are subsequences with a start and end positions in the parent reference sequence.

Tricky concepts. SimpleAlign expects name,start,end to be 'unique' in the alignment, and this is the key for the internal hashes. (name,start,end is abbreviated nse in the code). However, in many cases people don't want the name/start-end to be displayed: either multiple names in an alignment or names specific to the alignment (ROA1_HUMAN_1, ROA1_HUMAN_2 etc). These names are called 'displayname', and generally is what is used to print out the alignment. They default to name/start-end.

The SimpleAlign Module came from Ewan Birney's Align module.


SimpleAlign is being slowly converted to bioperl coding standards, mainly by Ewan.

Use Bio::Root::Object - done
Use proper exceptions - done
Use hashed constructor - not done!


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://bioperl.org/MailList.shtml - 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:



Ewan Birney, birney@sanger.ac.uk


David J. Evans, David.Evans@vir.gla.ac.uk Heikki Lehvaslaiho, heikki@ebi.ac.uk Jason Stajich, jason@bioperl.org Xintao Wei & Giri Narasimhan, giri@cs.fiu.edu




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


 Title     : new
 Usage     : my $aln = new Bio::SimpleAlign();
 Function  : Creates a new simple align object
 Returns   : Bio::SimpleAlign
 Args      : -source => string representing the source program 
                        where this alignment came from

Modifier methods

These methods modify the MSE by adding, removing or shuffling complete sequences.


 Title     : add_seq
 Usage     : $myalign->add_seq($newseq);
 Function  : Adds another sequence to the alignment. *Does not* align
             it - just adds it to the hashes.
 Returns   : nothing
 Args      : a Bio::LocatableSeq object
             order (optional)

See Bio::LocatableSeq for more information


 Title     : remove_seq
 Usage     : $aln->remove_seq($seq);
 Function  : Removes a single sequence from an alignment
 Returns   :
 Argument  : a Bio::LocatableSeq object


 Title   : purge
 Usage   : $aln->purge(0.7);

           Removes sequences above whatever %id.

           This function will grind on large alignments. Beware!
           (perhaps not ideally implemented)

 Example :
 Returns : An array of the removed sequences
 Args    :


 Title     : sort_alphabetically
 Usage     : $ali->sort_alphabetically
 Function  : 

             Changes the order of the alignemnt to alphabetical on name 
             followed by numerical by number.

 Returns   : 
 Argument  : 

Sequence selection methods

Methods returning one or more sequences objects.


 Title     : each_seq
 Usage     : foreach $seq ( $align->each_seq() ) 
 Function  : Gets an array of Seq objects from the alignment
 Returns   : an array
 Argument  : 


 Title     : each_alphabetically
 Usage     : foreach $seq ( $ali->each_alphabetically() )
 Function  :

             Returns an array of sequence object sorted alphabetically 
             by name and then by start point.
             Does not change the order of the alignment

 Returns   : 
 Argument  : 


 Title     : each_seq_with_id
 Usage     : foreach $seq ( $align->each_seq_with_id() ) 
 Function  : 

             Gets an array of Seq objects from the
             alignment, the contents being those sequences
             with the given name (there may be more than one)

 Returns   : an array
 Argument  : a seq name


 Title     : get_seq_by_pos
 Usage     : $seq = $aln->get_seq_by_pos(3) # third sequence from the alignment
 Function  : 

             Gets a sequence based on its position in the alignment.
             Numbering starts from 1.  Sequence positions larger than
             no_sequences() will thow an error.

 Returns   : a Bio::LocatableSeq object
 Args      : positive integer for the sequence osition

Create new alignments

The result of these methods are horizontal or vertical subsets of the current MSE.


 Title     : select
 Usage     : $aln2 = $aln->select(1, 3) # three first sequences
 Function  : 

             Creates a new alignment from a continuous subset of
             sequences.  Numbering starts from 1.  Sequence positions
             larger than no_sequences() will thow an error.

 Returns   : a Bio::SimpleAlign object
 Args      : positive integer for the first sequence
             positive integer for the last sequence to include (optional)


 Title     : select_noncont
 Usage     : $aln2 = $aln->select_noncont(1, 3) # first and 3rd sequences
 Function  : 

             Creates a new alignment from a subset of
             sequences.  Numbering starts from 1.  Sequence positions
             larger than no_sequences() will thow an error.

 Returns   : a Bio::SimpleAlign object
 Args      : array of integers for the sequences


 Title     : slice
 Usage     : $aln2 = $aln->slice(20, 30)
 Function  : 

             Creates a slice from the alignment inclusive of start and
             end columns.  Sequences with no residues in the slice are
             excluded from the new alignment and a warning is printed.
             Slice beyond the length of the sequence does not do

 Returns   : a Bio::SimpleAlign object
 Args      : positive integer for start column 
             positive integer for end column 

Change sequences within the MSE

These methods affect characters in all sequences without changeing the alignment.


 Title     : map_chars
 Usage     : $ali->map_chars('\.','-')
 Function  : 

             Does a s/$arg1/$arg2/ on the sequences. Useful for gap

             Notice that the from (arg1) is interpretted as a regex,
             so be careful about quoting meta characters (eg
             $ali->map_chars('.','-') wont do what you want)

 Returns   : 
 Argument  : 'from' rexexp
             'to' string


 Title     : uppercase()
 Usage     : $ali->uppercase()
 Function  : Sets all the sequences to uppercase
 Returns   : 
 Argument  : 


 Title    : cigar_line()
 Usage    : $align->cigar_line()
 Function : Generates a "cigar" line for each sequence in the alignment
            The format is simply A-1,60;B-1,1:4,60;C-5,10:12,58
            where A,B,C,etc. are the sequence identifiers, and the numbers
            refer to conserved positions within the alignment
 Args     : none


 Title    : match_line()
 Usage    : $align->match_line()
 Function : Generates a match line - much like consensus string
            except that a line indicating the '*' for a match.
 Args     : (optional) Match line characters ('*' by default)
            (optional) Strong match char (':' by default)
            (optional) Weak match char ('.' by default)


 Title     : match()
 Usage     : $ali->match()
 Function  : 

             Goes through all columns and changes residues that are
             identical to residue in first sequence to match '.'
             character. Sets L<match_char>.

             USE WITH CARE: Most MSE formats do not support match
             characters in sequences, so this is mostly for output
             only. NEXUS format (L<Bio::AlignIO::nexus>) can handle

 Returns   : 1
 Argument  : a match character, optional, defaults to '.'


 Title     : unmatch()
 Usage     : $ali->unmatch()
 Function  : 

             Undoes the effect of method L<match>. Unsets L<match_char>.

 Returns   : 1
 Argument  : a match character, optional, defaults to '.'

MSE attibutes

Methods for setting and reading the MSE attributes.

Note that the methods defining character semantics depend on the user to set them sensibly. They are needed only by certain input/output methods. Unset them by setting to an empty string ('').


 Title     : id
 Usage     : $myalign->id("Ig")
 Function  : Gets/sets the id field of the alignment
 Returns   : An id string
 Argument  : An id string (optional)


 Title     : missing_char
 Usage     : $myalign->missing_char("?")
 Function  : Gets/sets the missing_char attribute of the alignment
             It is generally recommended to set it to 'n' or 'N' 
             for nucleotides and to 'X' for protein. 
 Returns   : An missing_char string,
 Argument  : An missing_char string (optional)


 Title     : match_char
 Usage     : $myalign->match_char('.')
 Function  : Gets/sets the match_char attribute of the alignment
 Returns   : An match_char string,
 Argument  : An match_char string (optional)


 Title     : gap_char
 Usage     : $myalign->gap_char('-')
 Function  : Gets/sets the gap_char attribute of the alignment
 Returns   : An gap_char string, defaults to '-'
 Argument  : An gap_char string (optional)


 Title   : symbol_chars
 Usage   : my @symbolchars = $aln->symbol_chars;
 Function: Returns all the seen symbols (other than gaps)
 Returns : array of characters that are the seen symbols
 Args    : boolean to include the gap/missing/match characters

Alignment descriptors

These read only methods describe the MSE in various ways.


 Title     : consensus_string
 Usage     : $str = $ali->consensus_string($threshold_percent)
 Function  : Makes a strict consensus 
 Returns   : 
 Argument  : Optional treshold ranging from 0 to 100.
             The consensus residue has to appear at least threshold %
             of the sequences at a given location, otherwise a '?'
             character will be placed at that location.
             (Default value = 0%)


 Title     : consensus_iupac
 Usage     : $str = $ali->consensus_iupac()
 Function  : 

             Makes a consensus using IUPAC ambiguity codes from DNA
             and RNA. The output is in upper case except when gaps in
             a column force output to be in lower case.

             Note that if your alignment sequences contain a lot of
             IUPAC ambiquity codes you often have to manually set
             alphabet.  L<Bio::PrimarySeq::_guess_type> thinks they
             indicate a protein sequence.

 Returns   : consensus string
 Argument  : none
 Throws    : on protein sequences


 Title     : is_flush
 Usage     : if( $ali->is_flush() )
 Function  : Tells you whether the alignment 
           : is flush, ie all of the same length
 Returns   : 1 or 0
 Argument  : 


 Title     : length()
 Usage     : $len = $ali->length() 
 Function  : Returns the maximum length of the alignment.
             To be sure the alignment is a block, use is_flush
 Returns   : 
 Argument  : 


 Title     : maxdisplayname_length
 Usage     : $ali->maxdisplayname_length()
 Function  : 

             Gets the maximum length of the displayname in the
             alignment. Used in writing out various MSE formats.

 Returns   : integer
 Argument  : 


 Title     : no_residues
 Usage     : $no = $ali->no_residues
 Function  : number of residues in total in the alignment
 Returns   : integer
 Argument  : 


 Title     : no_sequences
 Usage     : $depth = $ali->no_sequences
 Function  : number of sequence in the sequence alignment
 Returns   : integer
 Argument  : 


 Title   : average_percentage_identity
 Usage   : $id = $align->average_percentage_identity
 Function: The function uses a fast method to calculate the average 
           percentage identity of the alignment
 Returns : The average percentage identity of the alignment
 Args    : None


 Title   : percentage_identity
 Usage   : $id = $align->percentage_identity
 Function: The function calculates the average percentage identity
           (aliased for average_percentage_identity)
 Returns : The average percentage identity 
 Args    : None


 Title   : percentage_identity
 Usage   : $id = $align->percentage_identity
 Function: The function calculates the percentage identity of 
           the conserved columns
 Returns : The percentage identity of the conserved columns
 Args    : None

Alignment positions

Methods to map a sequence position into an alignment column and back. column_from_residue_number() does the former. The latter is really a property of the sequence object and can done using Bio::LocatableSeq::location_from_column:

    # select somehow a sequence from the alignment, e.g.
    my $seq = $aln->get_seq_by_pos(1);
    #$loc is undef or Bio::LocationI object
    my $loc = $seq->location_from_column(5);


 Title   : column_from_residue_number
 Usage   : $col = $ali->column_from_residue_number( $seqname, $resnumber)

           This function gives the position in the alignment
           (i.e. column number) of the given residue number in the
           sequence with the given name. For example, for the

             Seq1/91-97 AC..DEF.GH
             Seq2/24-30 ACGG.RTY..
             Seq3/43-51 AC.DDEFGHI

           column_from_residue_number( "Seq1", 94 ) returns 5.
           column_from_residue_number( "Seq2", 25 ) returns 2.
           column_from_residue_number( "Seq3", 50 ) returns 9.

           An exception is thrown if the residue number would lie
           outside the length of the aligment
           (e.g. column_from_residue_number( "Seq2", 22 )

          Note: If the the parent sequence is represented by more than
          one alignment sequence and the residue number is present in
          them, this method finds only the first one.

 Returns : A column number for the position in the alignment of the
           given residue in the given sequence (1 = first column)
 Args    : A sequence id/name (not a name/start-end)
           A residue number in the whole sequence (not just that
           segment of it in the alignment)

Sequence names

Methods to manipulate the display name. The default name based on the sequence id and subsequence positions can be overridden in various ways.


 Title     : displayname
 Usage     : $myalign->displayname("Ig", "IgA")
 Function  : Gets/sets the display name of a sequence in the alignment
 Returns   : A display name string
 Argument  : name of the sequence
             displayname of the sequence (optional)


 Title     : set_displayname_count
 Usage     : $ali->set_displayname_count
 Function  : 

             Sets the names to be name_# where # is the number of
             times this name has been used.

 Returns   : 
 Argument  : 


 Title     : set_displayname_flat
 Usage     : $ali->set_displayname_flat()
 Function  : Makes all the sequences be displayed as just their name,
             not name/start-end
 Returns   : 1
 Argument  : 


 Title     : set_displayname_normal
 Usage     : $ali->set_displayname_normal() 
 Function  : Makes all the sequences be displayed as name/start-end
 Returns   : 
 Argument  : 


 Title   : source
 Usage   : $obj->source($newval)
 Function: sets the Alignment source program
 Example : 
 Returns : value of source
 Args    : newvalue (optional)