Transcript.pm

=head1 LICENSE

Copyright [1999-2015] Wellcome Trust Sanger Institute and the EMBL-European Bioinformatics Institute
Copyright [2016-2019] EMBL-European Bioinformatics Institute

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

     http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

=cut


=head1 CONTACT

  Please email comments or questions to the public Ensembl
  developers list at <http://lists.ensembl.org/mailman/listinfo/dev>.

  Questions may also be sent to the Ensembl help desk at
  <http://www.ensembl.org/Help/Contact>.

=cut

=head1 NAME

Bio::EnsEMBL::Transcript - object representing an Ensembl transcript

=head1 SYNOPSIS

Creation:

  my $tran = new Bio::EnsEMBL::Transcript();
  my $tran = new Bio::EnsEMBL::Transcript( -EXONS => \@exons );

Manipulation:

  # Returns an array of Exon objects
  my @exons = @{ $tran->get_all_Exons() };

  # Returns the peptide translation of the exons as a Bio::Seq
  if ( $tran->translation() ) {
    my $pep = $tran->translate();
  } else {
    print "Transcript ", $tran->stable_id(), " is non-coding\n";
  }

=head1 DESCRIPTION

A representation of a transcript within the Ensembl system.  A transcript
consists of a set of Exons and (possibly) a Translation which defines the
coding and non-coding regions of the exons.

=cut

package Bio::EnsEMBL::Transcript;

use strict;

use Bio::EnsEMBL::Feature;
use Bio::EnsEMBL::UTR;
use Bio::EnsEMBL::Intron;
use Bio::EnsEMBL::ExonTranscript;
use Bio::EnsEMBL::CDS;
use Bio::EnsEMBL::TranscriptMapper;
use Bio::EnsEMBL::SeqEdit;
use Bio::EnsEMBL::Biotype;
use Bio::EnsEMBL::Utils::Argument qw( rearrange );
use Bio::EnsEMBL::Utils::Exception qw(warning throw );
use Bio::EnsEMBL::Utils::Scalar qw( assert_ref );

use parent qw(Bio::EnsEMBL::Feature);

use constant SEQUENCE_ONTOLOGY => {
  acc  => 'SO:0000673',
  term => 'transcript',
};

=head2 new

  Arg [-EXONS] :
        reference to list of Bio::EnsEMBL::Exon objects - exons which make up 
        this transcript
  Arg [-STABLE_ID] :
        string - the stable identifier of this transcript
  Arg [-VERSION] :
        int - the version of the stable identifier of this transcript
  Arg [-EXTERNAL_NAME] :
        string - the external database name associated with this transcript
  Arg [-EXTERNAL_DB] :
        string - the name of the database the external name is from
  Arg [-EXTERNAL_STATUS]:
        string - the status of the external identifier
  Arg [-DISPLAY_XREF]:
        Bio::EnsEMBL::DBEntry - The external database entry that is used
        to label this transcript when it is displayed.
  Arg [-CREATED_DATE]:
        string - the date the transcript was created
  Arg [-MODIFIED_DATE]:
        string - the date the transcript was last modified
  Arg [-DESCRIPTION]:
        string - the transcripts description
  Arg [-BIOTYPE]: 
        string - the biotype e.g. "protein_coding"
  Arg [-IS_CURRENT]:
        Boolean - specifies if this is the current version of the transcript
  Arg [-SOURCE]:
        string - the transcript source, e.g. "ensembl"

  Example    : $tran = new Bio::EnsEMBL::Transcript(-EXONS => \@exons);
  Description: Constructor. Instantiates a Transcript object.
  Returntype : Bio::EnsEMBL::Transcript
  Exceptions : throw on bad arguments
  Caller     : general
  Status     : Stable

=cut

sub new {
  my $proto = shift;

  my $class = ref($proto) || $proto;

  my $self = $class->SUPER::new(@_);

  my (
    $exons,            $stable_id,    $version,
    $external_name,    $external_db,  $external_status,
    $display_xref,     $created_date, $modified_date,
    $description,      $biotype,      $confidence,
    $external_db_name, $is_current,
    $source
  );

    (
      $exons,            $stable_id,    $version,
      $external_name,    $external_db,  $external_status,
      $display_xref,     $created_date, $modified_date,
      $description,      $biotype,      $confidence,
      $external_db_name, $is_current,
      $source
      )
      = rearrange( [
        'EXONS',            'STABLE_ID',
        'VERSION',          'EXTERNAL_NAME',
        'EXTERNAL_DB',      'EXTERNAL_STATUS',
        'DISPLAY_XREF',     'CREATED_DATE',
        'MODIFIED_DATE',    'DESCRIPTION',
        'BIOTYPE',          'CONFIDENCE',
        'EXTERNAL_DB_NAME',
        'IS_CURRENT',       'SOURCE'
      ],
      @_
      );

  if ($exons) {
    $self->{'_trans_exon_array'} = $exons;
    $self->recalculate_coordinates();
  }

  $self->stable_id($stable_id);
  $self->{'created_date'}  = $created_date;
  $self->{'modified_date'} = $modified_date;
  $self->external_name($external_name) if ( defined $external_name );
  $self->external_db($external_db)     if ( defined $external_db );
  $self->external_status($external_status)
    if ( defined $external_status );
  $self->display_xref($display_xref) if ( defined $display_xref );
  $self->edits_enabled(1);

  $self->description($description);

  $self->{'biotype'} = $biotype;

  $self->source($source);

  # Default version
  if ( !defined($version) ) { $version = 1 }
  $self->{'version'} = $version;

  # default is_current
  $is_current = 1 unless ( defined($is_current) );
  $self->{'is_current'} = $is_current;

  return $self;
} ## end sub new

=head2 get_all_DBLinks

  Arg [1]    : String database name (optional)
               SQL wildcard characters (_ and %) can be used to
               specify patterns.

  Arg [2]    : (optional) String, external database type, can be one of
               ('ARRAY','ALT_TRANS','ALT_GENE','MISC','LIT','PRIMARY_DB_SYNONYM','ENSEMBL'),
               SQL wildcard characters (_ and %) can be used to
               specify patterns.

  Example    : my @dblinks = @{ $transcript->get_all_DBLinks() };
               @dblinks = @{ $transcript->get_all_DBLinks('Uniprot%') };}
               @dblinks = @{ $transcript->get_all_DBLinks('%', 'ENSEMBL') };

  Description: Retrieves *all* related DBEntries for this
               transcript.  This includes all DBEntries that are
               associated with the corresponding translation.

               If you only want to retrieve the DBEntries associated
               with the transcript (and not the translation) then
               you should use the get_all_DBEntries() call instead.

               Note: Each entry may be listed more than once.  No
               uniqueness checks are done.  Also if you put in an
               incorrect external database name no checks are done
               to see if this exists, you will just get an empty
               list.

  Return type: Listref of Bio::EnsEMBL::DBEntry objects
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub get_all_DBLinks {
  my ( $self, $db_name_exp, $ex_db_type ) = @_;

  my @links =
    @{ $self->get_all_DBEntries( $db_name_exp, $ex_db_type ) };

  # Add all of the transcript and translation xrefs to the return list.
  my $translation = $self->translation();
  if ( defined($translation) ) {
    push( @links,
          @{$translation->get_all_DBEntries( $db_name_exp, $ex_db_type ) }
    );
  }

  @links = sort { _compare_xrefs() } @links;

  return \@links;
}

=head2 get_all_xrefs

  Arg [1]    : String database name (optional)
               SQL wildcard characters (_ and %) can be used to
               specify patterns.

  Example    : @xrefs = @{ $transcript->get_all_xrefs() };
               @xrefs = @{ $transcript->get_all_xrefs('Uniprot%') };

  Description: Retrieves *all* related xrefs for this transcript.
               This includes all xrefs that are associated with the
               corresponding translation of this transcript.

               If you want to retrieve the xrefs associated with
               only the transcript (and not the translation) then
               you should use the get_all_object_xrefs() method
               instead.

               Note: Each entry may be listed more than once.  No
               uniqueness checks are done.  Also if you put in an
               incorrect external database name no checks are done
               to see if this exists, you will just get an empty
               list.

                NB: This method is an alias for the
                    get_all_DBLinks() method.

  Return type: Listref of Bio::EnsEMBL::DBEntry objects

  Status     : Stable

=cut

sub get_all_xrefs {
  my $self = shift;
  return $self->get_all_DBLinks(@_);
}

=head2 get_all_DBEntries

  Arg [1]    : (optional) String, external database name,
               SQL wildcard characters (_ and %) can be used to
               specify patterns.

  Arg [2]    : (optional) String, external database type, can be one of
               ('ARRAY','ALT_TRANS','ALT_GENE','MISC','LIT','PRIMARY_DB_SYNONYM','ENSEMBL'),
               SQL wildcard characters (_ and %) can be used to
               specify patterns.

  Example    : my @dbentries = @{ $transcript->get_all_DBEntries() };
               @dbentries = @{ $transcript->get_all_DBEntries('Uniprot%') };}
               @dbentries = @{ $transcript->get_all_DBEntries('%', 'ENSEMBL') };}

  Description: Retrieves DBEntries (xrefs) for this transcript.
               This does *not* include the corresponding
               translations DBEntries (see get_all_DBLinks()).

               This method will attempt to lazy-load DBEntries
               from a database if an adaptor is available and no
               DBEntries are present on the transcript (i.e. they
               have not already been added or loaded).

  Returntype : Listref of Bio::EnsEMBL::DBEntry objects
  Exceptions : none
  Caller     : get_all_DBLinks, TranscriptAdaptor::store
  Status     : Stable

=cut

sub get_all_DBEntries {
  my ( $self, $ex_db_exp, $ex_db_type ) = @_;

  my $cache_name = 'dbentries';

  if ( defined($ex_db_exp) ) {
    $cache_name .= $ex_db_exp;
  }

  if ( defined($ex_db_type) ) {
    $cache_name .= $ex_db_type;
  }

  # if not cached, retrieve all of the xrefs for this transcript
  if ( !defined( $self->{$cache_name} ) && defined( $self->adaptor() ) )
  {
    $self->{$cache_name} =
      $self->adaptor()->db()->get_DBEntryAdaptor()
      ->fetch_all_by_Transcript( $self, $ex_db_exp, $ex_db_type );
  }

  $self->{$cache_name} ||= [];

  return $self->{$cache_name};
} ## end sub get_all_DBEntries

=head2 get_all_object_xrefs

  Arg [1]    : (optional) String, external database name

  Arg [2]    : (optional) String, external_db type

  Example    : @oxrefs = @{ $transcript->get_all_object_xrefs() };

  Description: Retrieves xrefs for this transcript.  This does
               *not* include xrefs that are associated with the
               corresponding translations of this transcript (see
               get_all_xrefs()).

               This method will attempt to lazy-load xrefs from a
               database if an adaptor is available and no xrefs are
               present on the transcript (i.e. they have not already
               been added or loaded).

                NB: This method is an alias for the
                    get_all_DBentries() method.

  Return type: Listref of Bio::EnsEMBL::DBEntry objects

  Status     : Stable

=cut

sub get_all_object_xrefs {
  my $self = shift;
  return $self->get_all_DBEntries(@_);
}

=head2 add_DBEntry

  Arg [1]    : Bio::EnsEMBL::DBEntry $dbe
               The dbEntry to be added
  Example    : my $dbe = Bio::EnsEMBL::DBEntery->new(...);
               $transcript->add_DBEntry($dbe);
  Description: Associates a DBEntry with this transcript. Note that adding
               DBEntries will prevent future lazy-loading of DBEntries for this
               gene (see get_all_DBEntries).
  Returntype : none
  Exceptions : thrown on incorrect argument type
  Caller     : general
  Status     : Stable

=cut

sub add_DBEntry {
  my $self = shift;
  my $dbe = shift;

  unless($dbe && ref($dbe) && $dbe->isa('Bio::EnsEMBL::DBEntry')) {
    throw('Expected DBEntry argument');
  }

  $self->{'dbentries'} ||= [];
  push @{$self->{'dbentries'}}, $dbe;
}


=head2 get_all_supporting_features

  Example    : my @evidence = @{ $transcript->get_all_supporting_features };
  Description: Retrieves any supporting features added manually by 
               calls to add_supporting_features.
  Returntype : Listref of Bio::EnsEMBL::FeaturePair objects
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub get_all_supporting_features {
  my $self = shift;

  if( !exists  $self->{_supporting_evidence} )  {
    if($self->adaptor) {
      my $tsfa = $self->adaptor->db->get_TranscriptSupportingFeatureAdaptor();
      $self->{_supporting_evidence} = $tsfa->fetch_all_by_Transcript($self);
    }
  }
  
  return $self->{_supporting_evidence} || [];
}


=head2 add_supporting_features

  Arg [1-N]  : Bio::EnsEMBL::FeaturePair $feature
               The supporting features to add
  Example    : $transcript->add_supporting_features(@features);
  Description: Adds a list of supporting features to this Transcript.
               The added features can be retieved by
               get_all_supporting_features().
  Returntype : none
  Exceptions : throw if any of the features are not FeaturePairs
               throw if any of the features are not in the same coordinate
               system as the Transcript
  Caller     : general
  Status     : Stable
 
=cut
 
sub add_supporting_features {
  my ($self, @features) = @_;

  return unless @features;
 
  $self->{_supporting_evidence} ||= [];
  
  # check whether this feature object has been added already
  FEATURE: foreach my $feature (@features) {

    if (!defined($feature) || ref($feature) eq "ARRAY") {
      throw("Element in transcript supporting features array is undefined or is an ARRAY for " . $self->dbID);
    }
    if (!$feature || !$feature->isa("Bio::EnsEMBL::FeaturePair")) {
      print "feature = " . $feature . "\n";
      throw("Supporting feat [$feature] not a " .
            "Bio::EnsEMBL::FeaturePair");
    } 
    
    if ((defined $self->slice() && defined $feature->slice())&&
      ( $self->slice()->name() ne $feature->slice()->name())){
      throw("Supporting feat not in same coord system as transcript\n" .
            "transcript is attached to [".$self->slice()->name()."]\n" .
            "feat is attached to [".$feature->slice()->name()."]");
    }

    foreach my $added_feature ( @{ $self->{_supporting_evidence} } ){
      # compare objects
      if ( $feature == $added_feature ){
  #this feature has already been added
  next FEATURE;
      }
    }
    
    #no duplicate was found, add the feature
    push(@{$self->{_supporting_evidence}}, $feature);
  }
}


=head2 flush_supporting_features

  Example     : $transcript->flush_supporting_features;
  Description : Removes all supporting evidence from the transcript.
  Return type : (Empty) listref
  Exceptions  : none
  Caller      : general
  Status      : Stable

=cut

sub flush_supporting_features {
  my $self = shift;
  $self->{'_supporting_evidence'} = [];
}


=head2 external_db

  Arg [1]    : (optional) String - name of external db to set
  Example    : $transcript->external_db('HGNC');
  Description: Getter/setter for attribute external_db. The db is the one that 
               belongs to the external_name.  
  Returntype : String
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub external_db {
  my ( $self, $ext_dbname ) = @_;

  if(defined $ext_dbname) { 
    return ( $self->{'external_db'} = $ext_dbname );
  }

  if( exists $self->{'external_db'} ) {
    return $self->{'external_db'};
  }

  my $display_xref = $self->display_xref();

  if( defined $display_xref ) {
    return $display_xref->dbname()
  } else {
    return undef;
  }
}


=head2 external_status

  Arg [1]    : (optional) String - status of the external db
  Example    : $transcript->external_status('KNOWNXREF');
  Description: Getter/setter for attribute external_status. The status of
               the external db of the one that belongs to the external_name.
  Returntype : String
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub external_status { 
  my ( $self, $ext_status ) = @_;

  if(defined $ext_status) {
    return ( $self->{'external_status'} = $ext_status );
  }

  if( exists $self->{'external_status'} ) {
    return $self->{'external_status'};
  }

  my $display_xref = $self->display_xref();

  if( defined $display_xref ) {
    return $display_xref->status()
  } else {
    return undef;
  }
}


=head2 external_name

  Arg [1]    : (optional) String - the external name to set
  Example    : $transcript->external_name('BRCA2-001');
  Description: Getter/setter for attribute external_name.
  Returntype : String or undef
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub external_name {
  my ($self, $ext_name) = @_;

  if(defined $ext_name) { 
    return ( $self->{'external_name'} = $ext_name );
  }

  if( exists $self->{'external_name'} ) {
    return $self->{'external_name'};
  }

  my $display_xref = $self->display_xref();

  if( defined $display_xref ) {
    return $display_xref->display_id()
  } else {
    return undef;
  }
}

=head2 source

  Arg [1]    : (optional) String - the source to set
  Example    : $transcript->source('ensembl');
  Description: Getter/setter for attribute source
  Returntype : String
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub source {
  my $self = shift;
  $self->{'source'} = shift if( @_ );
  return ( $self->{'source'} || "ensembl" );
}

=head2 display_xref

  Arg [1]    : (optional) Bio::EnsEMBL::DBEntry - the display xref to set
  Example    : $transcript->display_xref($db_entry);
  Description: Getter/setter for display_xref for this transcript.
  Returntype : Bio::EnsEMBL::DBEntry
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub display_xref {
  my $self = shift;
  $self->{'display_xref'} = shift if(@_);
  return $self->{'display_xref'};
}

=head2 is_canonical

  Args [1]      : (optional) Boolean is_canonical

  Example       : if ($transcript->is_canonical()) { ... }

  Description : Returns true (non-zero) if the transcript is the
                canonical transcript of its gene, false (0) if not. If the code
                returns an undefined it is because its state is not currently
                known. Internally the code will consult the database for this
                value if it is unknown and the transcript has a dbID and an
                attached adaptor

  Return type   : Boolean

  Status        : Stable

=cut

sub is_canonical {
  my ( $self, $value ) = @_;
  
  #Shortcut call
  return $self->{is_canonical} if defined $self->{is_canonical};
  
  if ( defined($value) ) {
    $self->{is_canonical} = ( $value ? 1 : 0 );
  }
  else {
    if(! defined $self->{is_canonical} && $self->dbID() && $self->adaptor()) {
      $self->{is_canonical} = $self->adaptor()->is_Transcript_canonical($self);
    }
  }

  return $self->{is_canonical};
}

=head2 translation

  Args       : None
  Example    : if ( $transcript->translation() ) {
                 print( $transcript->translation()->stable_id(), "\n" );
               } else {
                 print("Pseudogene\n");
               }
  Description: Getter/setter for the Translation object which
               defines the CDS (and as a result the peptide encoded
               by) this transcript.  This function will return
               undef if this transcript is a pseudogene, i.e. a
               non-translating transcript such as an ncRNA.  This
               is the accepted method of determining whether a
               transcript is a pseudogene or not.
  Returntype : Bio::EnsEMBL::Translation
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub translation {
  my ( $self, $translation ) = @_;

  if ( defined($translation) ) {
    assert_ref( $translation, 'Bio::EnsEMBL::Translation' );

    $self->{'translation'} = $translation;
    $translation->transcript($self);

    $self->{'cdna_coding_start'} = undef;
    $self->{'cdna_coding_end'}   = undef;

    $self->{'coding_region_start'} = undef;
    $self->{'coding_region_end'}   = undef;

    $self->{'transcript_mapper'} = undef;

  } elsif ( @_ > 1 ) {
    if ( defined( $self->{'translation'} ) ) {
      # Removing existing translation

      $self->{'translation'}->transcript(undef);
      delete( $self->{'translation'} );

      $self->{'cdna_coding_start'} = undef;
      $self->{'cdna_coding_end'}   = undef;

      $self->{'coding_region_start'} = undef;
      $self->{'coding_region_end'}   = undef;

      $self->{'transcript_mapper'} = undef;
    }
  } elsif (   !exists( $self->{'translation'} )
            && defined( $self->adaptor() ) )
  {
    $self->{'translation'} =
      $self->adaptor()->db()->get_TranslationAdaptor()
      ->fetch_by_Transcript($self);
  }

  return $self->{'translation'};
} ## end sub translation

=head2 get_all_alternative_translations

  Args       : None
  Example    :

    my @alt_translations =
      @{ $transcript->get_all_alternative_translations() };

  Description:  Fetches all alternative translations defined for this
                transcript.  The canonical translation is not returned.

  Returntype : Arrayref to Bio::EnsEMBL::Translation
  Exceptions : None
  Caller     : General
  Status     : Stable

=cut

sub get_all_alternative_translations {
  my ($self) = @_;

  if (   !exists( $self->{'alternative_translations'} )
       && defined( $self->adaptor() ) )
  {
    my $pa = $self->adaptor()->db()->get_TranslationAdaptor();
    my @translations =
      @{ $pa->fetch_all_alternative_by_Transcript($self) };

    $self->{'alternative_translations'} = \@translations;
  }

  return $self->{'alternative_translations'};
}

=head2 add_alternative_translation

  Args       : Bio::EnsEMBL::Translation $translation
  Example    :

    $transcript->add_alternative_translation($translation);

  Description: Adds an alternative translation to this transcript.
  Returntype : None
  Exceptions : None
  Caller     : General
  Status     : Stable

=cut

sub add_alternative_translation {
  my ( $self, $translation ) = @_;

  if ( !(    defined($translation)
          && ref($translation)
          && $translation->isa('Bio::EnsEMBL::Translation') ) )
  {
    throw("Bio::EnsEMBL::Translation argument expected.");
  }

  # Load the existsing alternative translations from the database if
  # they haven't already been loaded.
  $self->get_all_alternative_translations();

  push( @{ $self->{'alternative_translations'} }, $translation );
}

=head2 spliced_seq

  Args       : soft_mask (opt)
               if specified, will return a sequence where UTR regions are lowercased
  Description: Retrieves all Exon sequences and concats them together.
               No phase padding magic is done, even if phases do not align.
  Returntype : Text
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub spliced_seq {
  my ( $self, $soft_mask ) = @_;

  my $seq_string = "";
  for my $ex ( @{$self->get_all_Exons()} ) {
    my $seq = $ex->seq();
    if(!$seq) {
      warning("Could not obtain seq for exon.  Transcript sequence may not " .
              "be correct.");
      $seq_string .= 'N' x $ex->length();
    } else {
      my $exon_seq = $seq->seq();
      if ($soft_mask) {
        my $padstr;
        if (!defined ($ex->coding_region_start($self))) {
          $exon_seq = lc($exon_seq);
        }

        if ($ex->coding_region_start($self) > $ex->start()) {
          my $forward_length = $ex->coding_region_start($self) - $ex->start();
          my $reverse_length = $ex->end() - $ex->coding_region_start($self);
          if ($ex->strand == 1) {
            $exon_seq = lc (substr($exon_seq, 0, $forward_length)) . substr($exon_seq, $forward_length); 
          } else {
            $exon_seq = substr($exon_seq, 0, $reverse_length+1) . lc(substr($exon_seq, $reverse_length+1));
          }
        }

        if ($ex->coding_region_end($self) < $ex->end()) {
          my $forward_length = $ex->coding_region_end($self) - $ex->start();
          my $reverse_length = $ex->end() - $ex->coding_region_end($self);
          if ($ex->strand == 1) {
            $exon_seq = substr($exon_seq, 0, $forward_length+1) . lc(substr($exon_seq, $forward_length+1));
          } else {
            $exon_seq = lc(substr($exon_seq, 0, $reverse_length)) . substr($exon_seq, $reverse_length);
          }
        }
      }
      $seq_string .= $exon_seq;
    }
  }

  # apply post transcriptional edits
  if($self->edits_enabled()) {
    my @seqeds = @{$self->get_all_SeqEdits()};

    # sort edits in reverse order to remove complication of
    # adjusting downstream edits
    @seqeds = sort {$b->start() <=> $a->start()} @seqeds;

    foreach my $se (@seqeds) {
      $se->apply_edit(\$seq_string);
    }
  }

  return $seq_string;
}


=head2 translateable_seq

  Args       : none
  Example    : print $transcript->translateable_seq(), "\n";
  Description: Returns a sequence string which is the the translateable part
               of the transcripts sequence.  This is formed by splicing all
               Exon sequences together and apply all defined RNA edits.
               Then the coding part of the sequence is extracted and returned.
               The code will not support monkey exons any more. If you want to
               have non phase matching exons, defined appropriate _rna_edit
               attributes!

               An empty string is returned if this transcript is a pseudogene
               (i.e. is non-translateable).
  Returntype : Text
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub translateable_seq {
  my ( $self ) = @_;

  if ( !$self->translation() ) {
    return '';
  }

  my $mrna = $self->spliced_seq();

  my $start = $self->cdna_coding_start();
  my $end   = $self->cdna_coding_end();

  $mrna = substr( $mrna, $start - 1, $end - $start + 1 );

  my $start_phase = $self->translation->start_Exon->phase();
  if( $start_phase > 0 ) {
    $mrna = "N"x$start_phase . $mrna;
  }
  if( ! $start || ! $end ) {
    return "";
  }

  return $mrna;
}


=head2 cdna_coding_start

  Arg [1]    : (optional) $value
  Example    : $relative_coding_start = $transcript->cdna_coding_start;
  Description: Retrieves the position of the coding start of this transcript
               in cdna coordinates (relative to the start of the 5prime end of
               the transcript, excluding introns, including utrs).

               This will return undef if this is a pseudogene (i.e. a
               transcript with no translation).
  Returntype : int
  Exceptions : none
  Caller     : five_prime_utr, get_all_snps, general
  Status     : Stable

=cut

sub cdna_coding_start {
  my $self = shift;

  if( @_ ) {
    $self->{'cdna_coding_start'} = shift;
  }

  if(!defined $self->{'cdna_coding_start'} && defined $self->translation){
    # calc coding start relative from the start of translation (in cdna coords)
    my $start = 0;

    my @exons = @{$self->get_all_Exons};
    my $exon;

    while($exon = shift @exons) {
      if($exon == $self->translation->start_Exon) {
        #add the utr portion of the start exon
        $start += $self->translation->start;
        last;
      } else {
        #add the entire length of this non-coding exon
        $start += $exon->length;
      }
    }

    # adjust cdna coords if sequence edits are enabled
    if($self->edits_enabled()) {
      my @seqeds = @{$self->get_all_SeqEdits()};
      if (scalar @seqeds) {
        my $transl_start = $self->get_all_Attributes('_transl_start');
        if (@{$transl_start}) {
          $start = $transl_start->[0]->value;
        } else {
          # sort in reverse order to avoid adjustment of downstream edits
          @seqeds = sort {$b->start() <=> $a->start()} @seqeds;

          foreach my $se (@seqeds) {
            # use less than start so that start of CDS can be extended
            if($se->start() < $start) {
              $start += $se->length_diff();
            }
          }
        }
      }
    }

    $self->{'cdna_coding_start'} = $start;
  }

  return $self->{'cdna_coding_start'};
}


=head2 cdna_coding_end

  Arg [1]    : (optional) $value
  Example    : $cdna_coding_end = $transcript->cdna_coding_end;
  Description: Retrieves the end of the coding region of this transcript in
               cdna coordinates (relative to the five prime end of the
               transcript, excluding introns, including utrs).

               This will return undef if this transcript is a pseudogene
               (i.e. a transcript with no translation and therefor no CDS).
  Returntype : int
  Exceptions : none
  Caller     : general
  Status     : Stable

=cut

sub cdna_coding_end {
  my $self = shift;

  if( @_ ) {
    $self->{'cdna_coding_end'} = shift;
  }

  if(!defined $self->{'cdna_coding_end'} && defined $self->translation) {
    my @exons = @{$self->get_all_Exons};

    my $end = 0;
    while(my $exon = shift @exons) {
      if($exon == $self->translation->end_Exon) {
        # add coding portion of the final coding exon
        $end += $self->translation->end;
        last;
      } else {
        # add entire exon
        $end += $exon->length;
      }
    }

    # adjust cdna coords if sequence edits are enabled
    if($self->edits_enabled()) {
      my @seqeds = @{$self->get_all_SeqEdits()};
      if (scalar @seqeds) {
        my $transl_end = $self->get_all_Attributes('_transl_end');
        if (@{$transl_end}) {
          $end = $transl_end->[0]->value;
        } else {
          # sort in reverse order to avoid adjustment of downstream edits
          @seqeds = sort {$b->start() <=> $a->start()} @seqeds;

          foreach my $se (@seqeds) {
            # use less than or equal to end+1 so end of the CDS can be extended
            if($se->start() <= $end + 1) {
              $end += $se->length_diff();
            }
          }
        }
      }
    }

    $self->{'cdna_coding_end'} = $end;
  }

  return $self->{'cdna_coding_end'};
}