Remove unused
[bioperl-live.git] / Bio / Seq.pm
blobd1c6b1610d0631e8d0832bbea2474a69dae8d40c
2 # BioPerl module for Bio::Seq
4 # Please direct questions and support issues to <bioperl-l@bioperl.org>
6 # Cared for by Ewan Birney <birney@ebi.ac.uk>
8 # Copyright Ewan Birney
10 # You may distribute this module under the same terms as perl itself
12 # POD documentation - main docs before the code
14 =head1 NAME
16 Bio::Seq - Sequence object, with features
18 =head1 SYNOPSIS
20 # This is the main sequence object in Bioperl
22 # gets a sequence from a file
23 $seqio = Bio::SeqIO->new( '-format' => 'embl' , -file => 'myfile.dat');
24 $seqobj = $seqio->next_seq();
26 # SeqIO can both read and write sequences; see Bio::SeqIO
27 # for more information and examples
29 # get from database
30 $db = Bio::DB::GenBank->new();
31 $seqobj = $db->get_Seq_by_acc('X78121');
33 # make from strings in script
34 $seqobj = Bio::Seq->new( -display_id => 'my_id',
35 -seq => $sequence_as_string);
37 # gets sequence as a string from sequence object
38 $seqstr = $seqobj->seq(); # actual sequence as a string
39 $seqstr = $seqobj->subseq(10,50); # slice in biological coordinates
41 # retrieves information from the sequence
42 # features must implement Bio::SeqFeatureI interface
44 @features = $seqobj->get_SeqFeatures(); # just top level
45 foreach my $feat ( @features ) {
46 print "Feature ",$feat->primary_tag," starts ",$feat->start," ends ",
47 $feat->end," strand ",$feat->strand,"\n";
49 # features retain link to underlying sequence object
50 print "Feature sequence is ",$feat->seq->seq(),"\n"
53 # sequences may have a species
55 if( defined $seq->species ) {
56 print "Sequence is from ",$species->binomial," [",$species->common_name,"]\n";
59 # annotation objects are Bio::AnnotationCollectionI's
60 $ann = $seqobj->annotation(); # annotation object
62 # references is one type of annotations to get. Also get
63 # comment and dblink. Look at Bio::AnnotationCollection for
64 # more information
66 foreach my $ref ( $ann->get_Annotations('reference') ) {
67 print "Reference ",$ref->title,"\n";
70 # you can get truncations, translations and reverse complements, these
71 # all give back Bio::Seq objects themselves, though currently with no
72 # features transfered
74 my $trunc = $seqobj->trunc(100,200);
75 my $rev = $seqobj->revcom();
77 # there are many options to translate - check out the docs
78 my $trans = $seqobj->translate();
80 # these functions can be chained together
82 my $trans_trunc_rev = $seqobj->trunc(100,200)->revcom->translate();
86 =head1 DESCRIPTION
88 A Seq object is a sequence with sequence features placed on it. The
89 Seq object contains a PrimarySeq object for the actual sequence and
90 also implements its interface.
92 In Bioperl we have 3 main players that people are going to use frequently
94 Bio::PrimarySeq - just the sequence and its names, nothing else.
95 Bio::SeqFeatureI - a feature on a sequence, potentially with a sequence
96 and a location and annotation.
97 Bio::Seq - A sequence and a collection of sequence features
98 (an aggregate) with its own annotation.
100 Although Bioperl is not tied heavily to file formats these distinctions do
101 map to file formats sensibly and for some bioinformaticians this might help
103 Bio::PrimarySeq - Fasta file of a sequence
104 Bio::SeqFeatureI - A single entry in an EMBL/GenBank/DDBJ feature table
105 Bio::Seq - A single EMBL/GenBank/DDBJ entry
107 By having this split we avoid a lot of nasty circular references
108 (sequence features can hold a reference to a sequence without the sequence
109 holding a reference to the sequence feature). See L<Bio::PrimarySeq> and
110 L<Bio::SeqFeatureI> for more information.
112 Ian Korf really helped in the design of the Seq and SeqFeature system.
114 =head2 Examples
116 A simple and fundamental block of code:
118 use Bio::SeqIO;
120 my $seqIOobj = Bio::SeqIO->new(-file=>"1.fa"); # create a SeqIO object
121 my $seqobj = $seqIOobj->next_seq; # get a Seq object
123 With the Seq object in hand one has access to a powerful set of Bioperl
124 methods and related Bioperl objects. This next script will take a file of sequences
125 in EMBL format and create a file of the reverse-complemented sequences
126 in Fasta format using Seq objects. It also prints out details about the
127 exons it finds as sequence features in Genbank Flat File format.
129 use Bio::Seq;
130 use Bio::SeqIO;
132 $seqin = Bio::SeqIO->new( -format => 'EMBL' , -file => 'myfile.dat');
133 $seqout= Bio::SeqIO->new( -format => 'Fasta', -file => '>output.fa');
135 while((my $seqobj = $seqin->next_seq())) {
136 print "Seen sequence ",$seqobj->display_id,", start of seq ",
137 substr($seqobj->seq,1,10),"\n";
138 if( $seqobj->alphabet eq 'dna') {
139 $rev = $seqobj->revcom;
140 $id = $seqobj->display_id();
141 $id = "$id.rev";
142 $rev->display_id($id);
143 $seqout->write_seq($rev);
146 foreach $feat ( $seqobj->get_SeqFeatures() ) {
147 if( $feat->primary_tag eq 'exon' ) {
148 print STDOUT "Location ",$feat->start,":",
149 $feat->end," GFF[",$feat->gff_string,"]\n";
154 Let's examine the script. The lines below import the Bioperl modules.
155 Seq is the main Bioperl sequence object and SeqIO is the Bioperl support
156 for reading sequences from files and to files
158 use Bio::Seq;
159 use Bio::SeqIO;
161 These two lines create two SeqIO streams: one for reading in sequences
162 and one for outputting sequences:
164 $seqin = Bio::SeqIO->new( -format => 'EMBL' , -file => 'myfile.dat');
165 $seqout= Bio::SeqIO->new( -format => 'Fasta', -file => '>output.fa');
167 Notice that in the "$seqout" case there is a greater-than sign,
168 indicating the file is being opened for writing.
170 Using the
172 '-argument' => value
174 syntax is common in Bioperl. The file argument is like an argument
175 to open() . You can also pass in filehandles or FileHandle objects by
176 using the -fh argument (see L<Bio::SeqIO> documentation for details).
177 Many formats in Bioperl are handled, including Fasta, EMBL, GenBank,
178 Swissprot (swiss), PIR, and GCG.
180 $seqin = Bio::SeqIO->new( -format => 'EMBL' , -file => 'myfile.dat');
181 $seqout= Bio::SeqIO->new( -format => 'Fasta', -file => '>output.fa');
183 This is the main loop which will loop progressively through sequences
184 in a file, and each call to $seqio-E<gt>next_seq() provides a new Seq
185 object from the file:
187 while((my $seqobj = $seqio->next_seq())) {
189 This print line below accesses fields in the Seq object directly. The
190 $seqobj-E<gt>display_id is the way to access the display_id attribute
191 of the Seq object. The $seqobj-E<gt>seq method gets the actual
192 sequence out as string. Then you can do manipulation of this if
193 you want to (there are however easy ways of doing truncation,
194 reverse-complement and translation).
196 print "Seen sequence ",$seqobj->display_id,", start of seq ",
197 substr($seqobj->seq,1,10),"\n";
199 Bioperl has to guess the alphabet of the sequence, being either 'dna',
200 'rna', or 'protein'. The alphabet attribute is one of these three
201 possibilities.
203 if( $seqobj->alphabet eq 'dna') {
205 The $seqobj-E<gt>revcom method provides the reverse complement of the Seq
206 object as another Seq object. Thus, the $rev variable is a reference to
207 another Seq object. For example, one could repeat the above print line
208 for this Seq object (putting $rev in place of $seqobj). In this
209 case we are going to output the object into the file stream we built
210 earlier on.
212 $rev = $seqobj->revcom;
214 When we output it, we want the id of the outputted object
215 to be changed to "$id.rev", ie, with .rev on the end of the name. The
216 following lines retrieve the id of the sequence object, add .rev
217 to this and then set the display_id of the rev sequence object to
218 this. Notice that to set the display_id attribute you just need
219 call the same method, display_id(), with the new value as an argument.
220 Getting and setting values with the same method is common in Bioperl.
222 $id = $seqobj->display_id();
223 $id = "$id.rev";
224 $rev->display_id($id);
226 The write_seq method on the SeqIO output object, $seqout, writes the
227 $rev object to the filestream we built at the top of the script.
228 The filestream knows that it is outputting in fasta format, and
229 so it provides fasta output.
231 $seqout->write_seq($rev);
233 This block of code loops over sequence features in the sequence
234 object, trying to find ones who have been tagged as 'exon'.
235 Features have start and end attributes and can be outputted
236 in Genbank Flat File format, GFF, a standarized format for sequence
237 features.
239 foreach $feat ( $seqobj->get_SeqFeatures() ) {
240 if( $feat->primary_tag eq 'exon' ) {
241 print STDOUT "Location ",$feat->start,":",
242 $feat->end," GFF[",$feat->gff_string,"]\n";
246 The code above shows how a few Bio::Seq methods suffice to read, parse,
247 reformat and analyze sequences from a file. A full list of methods
248 available to Bio::Seq objects is shown below. Bear in mind that some of
249 these methods come from PrimarySeq objects, which are simpler
250 than Seq objects, stripped of features (see L<Bio::PrimarySeq> for
251 more information).
253 # these methods return strings, and accept strings in some cases:
255 $seqobj->seq(); # string of sequence
256 $seqobj->subseq(5,10); # part of the sequence as a string
257 $seqobj->accession_number(); # when there, the accession number
258 $seqobj->alphabet(); # one of 'dna','rna',or 'protein'
259 $seqobj->version() # when there, the version
260 $seqobj->keywords(); # when there, the Keywords line
261 $seqobj->length() # length
262 $seqobj->desc(); # description
263 $seqobj->primary_id(); # a unique id for this sequence regardless
264 # of its display_id or accession number
265 $seqobj->display_id(); # the human readable id of the sequence
267 Some of these values map to fields in common formats. For example, The
268 display_id() method returns the LOCUS name of a Genbank entry,
269 the (\S+) following the E<gt> character in a Fasta file, the ID from
270 a SwissProt file, and so on. The desc() method will return the DEFINITION
271 line of a Genbank file, the description following the display_id in a
272 Fasta file, and the DE field in a SwissProt file.
274 # the following methods return new Seq objects, but
275 # do not transfer features across to the new object:
277 $seqobj->trunc(5,10) # truncation from 5 to 10 as new object
278 $seqobj->revcom # reverse complements sequence
279 $seqobj->translate # translation of the sequence
281 # if new() can be called this method returns 1, else 0
283 $seqobj->can_call_new
285 # the following method determines if the given string will be accepted
286 # by the seq() method - if the string is acceptable then validate()
287 # returns 1, or 0 if not
289 $seqobj->validate_seq($string)
291 # the following method returns or accepts a Species object:
293 $seqobj->species();
295 Please see L<Bio::Species> for more information on this object.
297 # the following method returns or accepts an Annotation object
298 # which in turn allows access to Annotation::Reference
299 # and Annotation::Comment objects:
301 $seqobj->annotation();
303 These annotations typically refer to entire sequences, unlike
304 features. See L<Bio::AnnotationCollectionI>,
305 L<Bio::Annotation::Collection>, L<Bio::Annotation::Reference>, and
306 L<Bio::Annotation::Comment> for details.
308 It is also important to be able to describe defined portions of a
309 sequence. The combination of some description and the corresponding
310 sub-sequence is called a feature - an exon and its coordinates within
311 a gene is an example of a feature, or a domain within a protein.
313 # the following methods return an array of SeqFeatureI objects:
315 $seqobj->get_SeqFeatures # The 'top level' sequence features
316 $seqobj->get_all_SeqFeatures # All sequence features, including sub-seq
317 # features, such as features in an exon
319 # to find out the number of features use:
321 $seqobj->feature_count
323 Here are just some of the methods available to SeqFeatureI objects:
325 # these methods return numbers:
327 $feat->start # start position (1 is the first base)
328 $feat->end # end position (2 is the second base)
329 $feat->strand # 1 means forward, -1 reverse, 0 not relevant
331 # these methods return or accept strings:
333 $feat->primary_tag # the name of the sequence feature, eg
334 # 'exon', 'glycoslyation site', 'TM domain'
335 $feat->source_tag # where the feature comes from, eg, 'EMBL_GenBank',
336 # or 'BLAST'
338 # this method returns the more austere PrimarySeq object, not a
339 # Seq object - the main difference is that PrimarySeq objects do not
340 # themselves contain sequence features
342 $feat->seq # the sequence between start,end on the
343 # correct strand of the sequence
345 See L<Bio::PrimarySeq> for more details on PrimarySeq objects.
347 # useful methods for feature comparisons, for start/end points
349 $feat->overlaps($other) # do $feat and $other overlap?
350 $feat->contains($other) # is $other completely within $feat?
351 $feat->equals($other) # do $feat and $other completely agree?
353 # one can also add features
355 $seqobj->add_SeqFeature($feat) # returns 1 if successful
357 # sub features. For complex join() statements, the feature
358 # is one sequence feature with many sub SeqFeatures
360 $feat->sub_SeqFeature # returns array of sub seq features
362 Please see L<Bio::SeqFeatureI> and L<Bio::SeqFeature::Generic>,
363 for more information on sequence features.
365 It is worth mentioning that one can also retrieve the start and end
366 positions of a feature using a Bio::LocationI object:
368 $location = $feat->location # $location is a Bio::LocationI object
369 $location->start; # start position
370 $location->end; # end position
372 This is useful because one needs a Bio::Location::SplitLocationI object
373 in order to retrieve the coordinates inside the Genbank or EMBL join()
374 statements (e.g. "CDS join(51..142,273..495,1346..1474)"):
376 if ( $feat->location->isa('Bio::Location::SplitLocationI') &&
377 $feat->primary_tag eq 'CDS' ) {
378 foreach $loc ( $feat->location->sub_Location ) {
379 print $loc->start . ".." . $loc->end . "\n";
383 See L<Bio::LocationI> and L<Bio::Location::SplitLocationI> for more
384 information.
386 =head1 Implemented Interfaces
388 This class implements the following interfaces.
390 =over 4
392 =item Bio::SeqI
394 Note that this includes implementing Bio::PrimarySeqI.
396 =item Bio::IdentifiableI
398 =item Bio::DescribableI
400 =item Bio::AnnotatableI
402 =item Bio::FeatureHolderI
404 =back
406 =head1 FEEDBACK
409 =head2 Mailing Lists
411 User feedback is an integral part of the evolution of this and other
412 Bioperl modules. Send your comments and suggestions preferably to one
413 of the Bioperl mailing lists. Your participation is much appreciated.
415 bioperl-l@bioperl.org - General discussion
416 http://bioperl.org/wiki/Mailing_lists - About the mailing lists
418 =head2 Support
420 Please direct usage questions or support issues to the mailing list:
422 I<bioperl-l@bioperl.org>
424 rather than to the module maintainer directly. Many experienced and
425 reponsive experts will be able look at the problem and quickly
426 address it. Please include a thorough description of the problem
427 with code and data examples if at all possible.
429 =head2 Reporting Bugs
431 Report bugs to the Bioperl bug tracking system to help us keep track
432 the bugs and their resolution. Bug reports can be submitted via the
433 web:
435 https://redmine.open-bio.org/projects/bioperl/
437 =head1 AUTHOR - Ewan Birney, inspired by Ian Korf objects
439 Email birney@ebi.ac.uk
441 =head1 CONTRIBUTORS
443 Jason Stajich E<lt>jason@bioperl.orgE<gt>
444 Mark A. Jensen maj -at- fortinbras -dot- us
446 =head1 APPENDIX
449 The rest of the documentation details each of the object
450 methods. Internal methods are usually preceded with a "_".
452 =cut
455 # Let the code begin...
458 package Bio::Seq;
459 use strict;
461 use Bio::Annotation::Collection;
462 use Bio::PrimarySeq;
464 use base qw(Bio::Root::Root Bio::SeqI Bio::IdentifiableI Bio::DescribableI Bio::AnnotatableI Bio::FeatureHolderI Bio::AnnotationCollectionI);
466 =head2 new
468 Title : new
469 Usage : $seq = Bio::Seq->new( -seq => 'ATGGGGGTGGTGGTACCCT',
470 -id => 'human_id',
471 -accession_number => 'AL000012',
474 Function: Returns a new Seq object from
475 basic constructors, being a string for the sequence
476 and strings for id and accession_number
477 Returns : a new Bio::Seq object
479 =cut
481 sub new {
482 my($caller,@args) = @_;
484 if( $caller ne 'Bio::Seq') {
485 $caller = ref($caller) if ref($caller);
488 # we know our inherietance hierarchy
489 my $self = Bio::Root::Root->new(@args);
490 bless $self,$caller;
492 # this is way too sneaky probably. We delegate the construction of
493 # the Seq object onto PrimarySeq and then pop primary_seq into
494 # our primary_seq slot
496 my $pseq = Bio::PrimarySeq->new(@args);
498 # as we have just made this, we know it is ok to set hash directly
499 # rather than going through the method
501 $self->{'primary_seq'} = $pseq;
503 # setting this array is now delayed until the final
504 # moment, again speed ups for non feature containing things
505 # $self->{'_as_feat'} = [];
508 my ($ann, $pid,$feat,$species) = &Bio::Root::RootI::_rearrange($self,[qw(ANNOTATION PRIMARY_ID FEATURES SPECIES)], @args);
510 # for a number of cases - reading fasta files - these are never set. This
511 # gives a quick optimisation around testing things later on
513 if( defined $ann || defined $pid || defined $feat || defined $species ) {
514 $pid && $self->primary_id($pid);
515 $species && $self->species($species);
516 $ann && $self->annotation($ann);
518 if( defined $feat ) {
519 if( ref($feat) !~ /ARRAY/i ) {
520 if( ref($feat) && $feat->isa('Bio::SeqFeatureI') ) {
521 $self->add_SeqFeature($feat);
522 } else {
523 $self->warn("Must specify a valid Bio::SeqFeatureI or ArrayRef of Bio::SeqFeatureI's with the -features init parameter for ".ref($self));
525 } else {
526 foreach my $feature ( @$feat ) {
527 $self->add_SeqFeature($feature);
533 return $self;
537 =head1 PrimarySeq interface
540 The PrimarySeq interface provides the basic sequence getting
541 and setting methods for on all sequences.
543 These methods implement the Bio::PrimarySeq interface by delegating
544 to the primary_seq inside the object. This means that you
545 can use a Seq object wherever there is a PrimarySeq, and
546 of course, you are free to use these functions anyway.
548 =cut
550 =head2 seq
552 Title : seq
553 Usage : $string = $obj->seq()
554 Function: Get/Set the sequence as a string of letters. The
555 case of the letters is left up to the implementer.
556 Suggested cases are upper case for proteins and lower case for
557 DNA sequence (IUPAC standard),
558 but implementations are suggested to keep an open mind about
559 case (some users... want mixed case!)
560 Returns : A scalar
561 Args : Optionally on set the new value (a string). An optional second
562 argument presets the alphabet (otherwise it will be guessed).
563 Both parameters may also be given in named parameter style
564 with -seq and -alphabet being the names.
566 =cut
568 sub seq {
569 return shift->primary_seq()->seq(@_);
573 =head2 validate_seq
575 Title : validate_seq
576 Usage : if(! $seqobj->validate_seq($seq_str) ) {
577 print "sequence $seq_str is not valid for an object of
578 alphabet ",$seqobj->alphabet, "\n";
580 Function: Test that the given sequence is valid, i.e. contains only valid
581 characters. The allowed characters are all letters (A-Z) and '-','.',
582 '*','?','=' and '~'. Spaces are not valid. Note that this
583 implementation does not take alphabet() into account.
584 Returns : 1 if the supplied sequence string is valid, 0 otherwise.
585 Args : - Sequence string to be validated
586 - Boolean to throw an error if the sequence is invalid
588 =cut
590 sub validate_seq {
591 return shift->primary_seq()->validate_seq(@_);
595 =head2 length
597 Title : length
598 Usage : $len = $seq->length()
599 Function:
600 Example :
601 Returns : Integer representing the length of the sequence.
602 Args : None
604 =cut
606 sub length {
607 return shift->primary_seq()->length(@_);
611 =head1 Methods from the Bio::PrimarySeqI interface
613 =head2 subseq
615 Title : subseq
616 Usage : $substring = $obj->subseq(10,40);
617 Function: Returns the subseq from start to end, where the first base
618 is 1 and the number is inclusive, ie 1-2 are the first two
619 bases of the sequence
621 Start cannot be larger than end but can be equal
623 Returns : A string
624 Args : 2 integers
627 =cut
629 sub subseq {
630 return shift->primary_seq()->subseq(@_);
634 =head2 display_id
636 Title : display_id
637 Usage : $id = $obj->display_id or $obj->display_id($newid);
638 Function: Gets or sets the display id, also known as the common name of
639 the Seq object.
641 The semantics of this is that it is the most likely string
642 to be used as an identifier of the sequence, and likely to
643 have "human" readability. The id is equivalent to the LOCUS
644 field of the GenBank/EMBL databanks and the ID field of the
645 Swissprot/sptrembl database. In fasta format, the >(\S+) is
646 presumed to be the id, though some people overload the id
647 to embed other information. Bioperl does not use any
648 embedded information in the ID field, and people are
649 encouraged to use other mechanisms (accession field for
650 example, or extending the sequence object) to solve this.
652 Notice that $seq->id() maps to this function, mainly for
653 legacy/convenience issues.
654 Returns : A string
655 Args : None or a new id
657 =cut
659 sub display_id {
660 return shift->primary_seq->display_id(@_);
664 =head2 accession_number
666 Title : accession_number
667 Usage : $unique_biological_key = $obj->accession_number;
668 Function: Returns the unique biological id for a sequence, commonly
669 called the accession_number. For sequences from established
670 databases, the implementors should try to use the correct
671 accession number. Notice that primary_id() provides the
672 unique id for the implemetation, allowing multiple objects
673 to have the same accession number in a particular implementation.
675 For sequences with no accession number, this method should return
676 "unknown".
678 Can also be used to set the accession number.
679 Example : $key = $seq->accession_number or $seq->accession_number($key)
680 Returns : A string
681 Args : None or an accession number
683 =cut
685 sub accession_number {
686 return shift->primary_seq->accession_number(@_);
690 =head2 desc
692 Title : desc
693 Usage : $seqobj->desc($string) or $seqobj->desc()
694 Function: Sets or gets the description of the sequence
695 Example :
696 Returns : The description
697 Args : The description or none
699 =cut
701 sub desc {
702 return shift->primary_seq->desc(@_);
706 =head2 primary_id
708 Title : primary_id
709 Usage : $unique_implementation_key = $obj->primary_id;
710 Function: Returns the unique id for this object in this
711 implementation. This allows implementations to manage
712 their own object ids in a way the implementation can control
713 clients can expect one id to map to one object.
715 For sequences with no natural id, this method should return
716 a stringified memory location.
718 Can also be used to set the primary_id (or unset to undef).
720 [Note this method name is likely to change in 1.3]
722 Example : $id = $seq->primary_id or $seq->primary_id($id)
723 Returns : A string
724 Args : None or an id, or undef to unset the primary id.
726 =cut
728 sub primary_id {
729 # Note: this used to not delegate to the primary seq. This is
730 # really bad in very subtle ways. E.g., if you created the object
731 # with a primary id given to the constructor and then later you
732 # change the primary id, if this method wouldn't delegate you'd
733 # have different values for primary id in the PrimarySeq object
734 # compared to this instance. Not good.
736 # I can't remember why not delegating was ever deemed
737 # advantageous, but I hereby claim that its problems far outweigh
738 # its advantages, if there are any. Convince me otherwise if you
739 # disagree. HL 2004/08/05
741 return shift->primary_seq->primary_id(@_);
745 =head2 can_call_new
747 Title : can_call_new
748 Usage : if ( $obj->can_call_new ) {
749 $newobj = $obj->new( %param );
751 Function: can_call_new returns 1 or 0 depending
752 on whether an implementation allows new
753 constructor to be called. If a new constructor
754 is allowed, then it should take the followed hashed
755 constructor list.
757 $myobject->new( -seq => $sequence_as_string,
758 -display_id => $id
759 -accession_number => $accession
760 -alphabet => 'dna',
762 Example :
763 Returns : 1 or 0
764 Args : None
766 =cut
768 sub can_call_new {
769 return 1;
773 =head2 alphabet
775 Title : alphabet
776 Usage : if ( $obj->alphabet eq 'dna' ) { /Do Something/ }
777 Function: Get/Set the type of sequence being one of
778 'dna', 'rna' or 'protein'. This is case sensitive.
780 This is not called <type> because this would cause
781 upgrade problems from the 0.5 and earlier Seq objects.
783 Returns : A string either 'dna','rna','protein'. NB - the object must
784 make a call of the type - if there is no type specified it
785 has to guess.
786 Args : optional string to set : 'dna' | 'rna' | 'protein'
788 =cut
790 sub alphabet {
791 my $self = shift;
792 return $self->primary_seq->alphabet(@_) if @_ && defined $_[0];
793 return $self->primary_seq->alphabet();
797 =head2 is_circular
799 Title : is_circular
800 Usage : if( $obj->is_circular) { /Do Something/ }
801 Function: Returns true if the molecule is circular
802 Returns : Boolean value
803 Args : none
805 =cut
807 sub is_circular {
808 return shift->primary_seq()->is_circular(@_);
812 =head1 Methods for Bio::IdentifiableI compliance
814 =head2 object_id
816 Title : object_id
817 Usage : $string = $obj->object_id()
818 Function: a string which represents the stable primary identifier
819 in this namespace of this object. For DNA sequences this
820 is its accession_number, similarly for protein sequences
822 This is aliased to accession_number().
823 Returns : A scalar
825 =cut
827 sub object_id {
828 return shift->accession_number(@_);
832 =head2 version
834 Title : version
835 Usage : $version = $obj->version()
836 Function: a number which differentiates between versions of
837 the same object. Higher numbers are considered to be
838 later and more relevant, but a single object described
839 the same identifier should represent the same concept
841 Returns : A number
843 =cut
845 sub version{
846 return shift->primary_seq->version(@_);
850 =head2 authority
852 Title : authority
853 Usage : $authority = $obj->authority()
854 Function: a string which represents the organisation which
855 granted the namespace, written as the DNS name for
856 organisation (eg, wormbase.org)
858 Returns : A scalar
860 =cut
862 sub authority {
863 return shift->primary_seq()->authority(@_);
867 =head2 namespace
869 Title : namespace
870 Usage : $string = $obj->namespace()
871 Function: A string representing the name space this identifier
872 is valid in, often the database name or the name
873 describing the collection
875 Returns : A scalar
877 =cut
879 sub namespace{
880 return shift->primary_seq()->namespace(@_);
884 =head1 Methods for Bio::DescribableI compliance
886 =head2 display_name
888 Title : display_name
889 Usage : $string = $obj->display_name()
890 Function: A string which is what should be displayed to the user
891 the string should have no spaces (ideally, though a cautious
892 user of this interface would not assumme this) and should be
893 less than thirty characters (though again, double checking
894 this is a good idea)
896 This is aliased to display_id().
897 Returns : A scalar
899 =cut
901 sub display_name {
902 return shift->display_id(@_);
905 =head2 description
907 Title : description
908 Usage : $string = $obj->description()
909 Function: A text string suitable for displaying to the user a
910 description. This string is likely to have spaces, but
911 should not have any newlines or formatting - just plain
912 text. The string should not be greater than 255 characters
913 and clients can feel justified at truncating strings at 255
914 characters for the purposes of display
916 This is aliased to desc().
917 Returns : A scalar
919 =cut
921 sub description {
922 return shift->desc(@_);
926 =head1 Methods for implementing Bio::AnnotatableI
928 =head2 annotation
930 Title : annotation
931 Usage : $ann = $seq->annotation or
932 $seq->annotation($ann)
933 Function: Gets or sets the annotation
934 Returns : Bio::AnnotationCollectionI object
935 Args : None or Bio::AnnotationCollectionI object
937 See L<Bio::AnnotationCollectionI> and L<Bio::Annotation::Collection>
938 for more information
940 =cut
942 sub annotation {
943 my ($obj,$value) = @_;
944 if( defined $value ) {
945 $obj->throw("object of class ".ref($value)." does not implement ".
946 "Bio::AnnotationCollectionI. Too bad.")
947 unless $value->isa("Bio::AnnotationCollectionI");
948 $obj->{'_annotation'} = $value;
949 } elsif( ! defined $obj->{'_annotation'}) {
950 $obj->{'_annotation'} = Bio::Annotation::Collection->new();
952 return $obj->{'_annotation'};
956 =head1 Methods for delegating Bio::AnnotationCollectionI
958 =head2 get_Annotations()
960 Usage : my @annotations = $seq->get_Annotations('key')
961 Function: Retrieves all the Bio::AnnotationI objects for a specific key
962 for this object
963 Returns : list of Bio::AnnotationI - empty if no objects stored for a key
964 Args : string which is key for annotations
966 =cut
968 sub get_Annotations { shift->annotation->get_Annotations(@_); }
971 =head2 add_Annotation()
973 Usage : $seq->add_Annotation('reference',$object);
974 $seq->add_Annotation($object,'Bio::MyInterface::DiseaseI');
975 $seq->add_Annotation($object);
976 $seq->add_Annotation('disease',$object,'Bio::MyInterface::DiseaseI');
977 Function: Adds an annotation for a specific key for this sequence object.
979 If the key is omitted, the object to be added must provide a value
980 via its tagname().
982 If the archetype is provided, this and future objects added under
983 that tag have to comply with the archetype and will be rejected
984 otherwise.
986 Returns : none
987 Args : annotation key ('disease', 'dblink', ...)
988 object to store (must be Bio::AnnotationI compliant)
989 [optional] object archetype to map future storage of object
990 of these types to
992 =cut
994 sub add_Annotation { shift->annotation->add_Annotation(@_) }
997 =head2 remove_Annotations()
999 Usage : $seq->remove_Annotations()
1000 Function: Remove the annotations for the specified key from this sequence
1001 object
1002 Returns : an list of Bio::AnnotationI compliant objects which were stored
1003 under the given key(s) for this sequence object
1004 Args : the key(s) (tag name(s), one or more strings) for which to
1005 remove annotations (optional; if none given, flushes all
1006 annotations)
1008 =cut
1010 sub remove_Annotations { shift->annotation->remove_Annotations(@_) }
1013 =head2 get_num_of_annotations()
1015 Usage : my $count = $seq->get_num_of_annotations()
1016 Alias : num_Annotations
1017 Function: Returns the count of all annotations stored for this sequence
1018 object
1019 Returns : integer
1020 Args : none
1022 =cut
1024 sub get_num_of_annotations { shift->annotation->get_num_of_annotations(@_) }
1025 sub num_Annotations { shift->get_num_of_annotations }; #DWYM
1028 =head1 Methods to implement Bio::FeatureHolderI
1030 This includes methods for retrieving, adding, and removing features.
1032 =cut
1034 =head2 get_SeqFeatures
1036 Title : get_SeqFeatures
1037 Usage :
1038 Function: Get the feature objects held by this feature holder.
1040 Features which are not top-level are subfeatures of one or
1041 more of the returned feature objects, which means that you
1042 must traverse the subfeature arrays of each top-level
1043 feature object in order to traverse all features associated
1044 with this sequence.
1046 Top-level features can be obtained by tag, specified in
1047 the argument.
1049 Use get_all_SeqFeatures() if you want the feature tree
1050 flattened into one single array.
1052 Example :
1053 Returns : an array of Bio::SeqFeatureI implementing objects
1054 Args : [optional] scalar string (feature tag)
1056 =cut
1058 sub get_SeqFeatures{
1059 my $self = shift;
1060 my $tag = shift;
1062 if( !defined $self->{'_as_feat'} ) {
1063 $self->{'_as_feat'} = [];
1065 if ($tag) {
1066 return map { $_->primary_tag eq $tag ? $_ : () } @{$self->{'_as_feat'}};
1068 else {
1069 return @{$self->{'_as_feat'}};
1074 =head2 get_all_SeqFeatures
1076 Title : get_all_SeqFeatures
1077 Usage : @feat_ary = $seq->get_all_SeqFeatures();
1078 Function: Returns the tree of feature objects attached to this
1079 sequence object flattened into one single array. Top-level
1080 features will still contain their subfeature-arrays, which
1081 means that you will encounter subfeatures twice if you
1082 traverse the subfeature tree of the returned objects.
1084 Use get_SeqFeatures() if you want the array to contain only
1085 the top-level features.
1087 Returns : An array of Bio::SeqFeatureI implementing objects.
1088 Args : None
1090 =cut
1092 # this implementation is inherited from FeatureHolderI
1094 =head2 feature_count
1096 Title : feature_count
1097 Usage : $seq->feature_count()
1098 Function: Return the number of SeqFeatures attached to a sequence
1099 Returns : integer representing the number of SeqFeatures
1100 Args : None
1102 =cut
1104 sub feature_count {
1105 my ($self) = @_;
1107 if (defined($self->{'_as_feat'})) {
1108 return ($#{$self->{'_as_feat'}} + 1);
1109 } else {
1110 return 0;
1115 =head2 add_SeqFeature
1117 Title : add_SeqFeature
1118 Usage : $seq->add_SeqFeature($feat);
1119 Function: Adds the given feature object to the feature array of this
1120 sequence. The object passed is required to implement the
1121 Bio::SeqFeatureI interface.
1122 The 'EXPAND' qualifier (see L<Bio::FeatureHolderI>) is supported, but
1123 has no effect,
1124 Returns : 1 on success
1125 Args : A Bio::SeqFeatureI implementing object.
1127 =cut
1129 sub add_SeqFeature {
1130 my ($self, @feat) = @_;
1132 $self->{'_as_feat'} = [] unless $self->{'_as_feat'};
1134 if (scalar @feat > 1) {
1135 $self->deprecated(
1136 -message => 'Providing an array of features to Bio::Seq add_SeqFeature()'.
1137 ' is deprecated and will be removed in a future version. '.
1138 'Add a single feature at a time instead.',
1139 -warn_version => 1.007,
1140 -throw_version => 1.009,
1144 for my $feat ( @feat ) {
1146 next if $feat eq 'EXPAND'; # Need to support it for FeatureHolderI compliance
1148 if( !$feat->isa("Bio::SeqFeatureI") ) {
1149 $self->throw("Expected a Bio::SeqFeatureI object, but got a $feat.");
1152 # make sure we attach ourselves to the feature if the feature wants it
1153 my $aseq = $self->primary_seq;
1154 $feat->attach_seq($aseq) if $aseq;
1156 push(@{$self->{'_as_feat'}},$feat);
1158 return 1;
1162 =head2 remove_SeqFeatures
1164 Title : remove_SeqFeatures
1165 Usage : $seq->remove_SeqFeatures();
1166 Function: Flushes all attached SeqFeatureI objects.
1168 To remove individual feature objects, delete those from the returned
1169 array and re-add the rest.
1170 Example :
1171 Returns : The array of Bio::SeqFeatureI objects removed from this seq.
1172 Args : None
1174 =cut
1176 sub remove_SeqFeatures {
1177 my $self = shift;
1179 return () unless $self->{'_as_feat'};
1180 my @feats = @{$self->{'_as_feat'}};
1181 $self->{'_as_feat'} = [];
1182 return @feats;
1186 =head1 Methods provided in the Bio::PrimarySeqI interface
1188 These methods are inherited from the PrimarySeq interface
1189 and work as one expects, building new Bio::Seq objects
1190 or other information as expected. See L<Bio::PrimarySeq>
1191 for more information.
1193 Sequence Features are B<not> transferred to the new objects.
1194 This is possibly a mistake. Anyone who feels the urge in
1195 dealing with this is welcome to give it a go.
1197 =head2 revcom
1199 Title : revcom
1200 Usage : $rev = $seq->revcom()
1201 Function: Produces a new Bio::Seq object which
1202 is the reversed complement of the sequence. For protein
1203 sequences this throws an exception of "Sequence is a protein.
1204 Cannot revcom"
1206 The id is the same id as the original sequence, and the
1207 accession number is also identical. If someone wants to track
1208 that this sequence has be reversed, it needs to define its own
1209 extensions
1211 To do an in-place edit of an object you can go:
1213 $seq = $seq->revcom();
1215 This of course, causes Perl to handle the garbage collection of
1216 the old object, but it is roughly speaking as efficient as an
1217 in-place edit.
1219 Returns : A new (fresh) Bio::Seq object
1220 Args : None
1222 =head2 trunc
1224 Title : trunc
1225 Usage : $subseq = $myseq->trunc(10,100);
1226 Function: Provides a truncation of a sequence
1228 Example :
1229 Returns : A fresh Seq object
1230 Args : A Seq object
1232 =head2 id
1234 Title : id
1235 Usage : $id = $seq->id()
1236 Function: This is mapped on display_id
1237 Returns : value of display_id()
1238 Args : [optional] value to update display_id
1240 =cut
1242 sub id {
1243 return shift->display_id(@_);
1247 =head1 Seq only methods
1249 These methods are specific to the Bio::Seq object, and not
1250 found on the Bio::PrimarySeq object
1252 =head2 primary_seq
1254 Title : primary_seq
1255 Usage : $seq->primary_seq or $seq->primary_seq($newval)
1256 Function: Get or set a PrimarySeq object
1257 Example :
1258 Returns : PrimarySeq object
1259 Args : None or PrimarySeq object
1261 =cut
1263 sub primary_seq {
1264 my ($obj,$value) = @_;
1266 if( defined $value) {
1267 if( ! ref $value || ! $value->isa('Bio::PrimarySeqI') ) {
1268 $obj->throw("$value is not a Bio::PrimarySeq compliant object");
1271 $obj->{'primary_seq'} = $value;
1272 # descend down over all seqfeature objects, seeing whether they
1273 # want an attached seq.
1275 foreach my $sf ( $obj->get_SeqFeatures() ) {
1276 $sf->attach_seq($value);
1280 return $obj->{'primary_seq'};
1285 =head2 species
1287 Title : species
1288 Usage : $species = $seq->species() or $seq->species($species)
1289 Function: Gets or sets the species
1290 Returns : L<Bio::Species> object
1291 Args : None or L<Bio::Species> object
1293 See L<Bio::Species> for more information
1295 =cut
1297 sub species {
1298 my ($self, $species) = @_;
1299 if ($species) {
1300 $self->{'species'} = $species;
1301 } else {
1302 return $self->{'species'};
1307 # Internal methods follow...
1309 # keep AUTOLOAD happy
1310 sub DESTROY { }
1312 ############################################################################
1313 # aliases due to name changes or to compensate for our lack of consistency #
1314 ############################################################################
1316 # in all other modules we use the object in the singular --
1317 # lack of consistency sucks
1318 *flush_SeqFeature = \&remove_SeqFeatures;
1319 *flush_SeqFeatures = \&remove_SeqFeatures;
1321 # this is now get_SeqFeatures() (from FeatureHolderI)
1322 *top_SeqFeatures = \&get_SeqFeatures;
1324 # this is now get_all_SeqFeatures() in FeatureHolderI
1325 sub all_SeqFeatures{
1326 return shift->get_all_SeqFeatures(@_);
1329 sub accession {
1330 my $self = shift;
1331 $self->warn(ref($self)."::accession is deprecated, ".
1332 "use accession_number() instead");
1333 return $self->accession_number(@_);