Bio::EnsEMBL::Analysis::Tools::Algorithms
TranscriptCluster
Toolbar
Summary
TranscriptCluster
Package variables
No package variables defined.
Included modules
Inherit
Synopsis
Description
This object holds one or more transcripts which have been clustered according to
comparison criteria external to this class (for instance, in the
method _compare_Transcripts of the class GeneComparison).
Each TranscriptCluster object holds the IDs of the transcripts clustered and the beginning and end coordinates
of each one (taken from the start and end coordinates of the first and last exon in the correspondig
get_all_Exons array)
It inherits from Bio::EnsEMBL::Root and Bio::RangeI. A TranscriptCluster is a range in the sense that it convers
a defined extent of genomic sequence. It is also possible to check whether two clusters overlap (in range),
is included into another cluster, etc...
Methods
Methods description
function to get the start and end positions - written as one method
for efficiency |
Title : end
Usage : $end = $transcript_cluster->end();
Function: get/set the end of the range covered by the cluster. This is re-calculated and set everytime
a new transcript is added to the cluster
Returns : the end of this range
Args : optionaly allows the end to be set
: using $range->end($end |
it returns the array of transcripts in the GeneCluster object |
Title : intersection
Usage : $intersection_cluster = $cluster1->intersection($cluster2)
Function: gives a cluster with the transcripts which fall entirely within the intersecting range of
$cluster1 and $cluster2
Args : arg #1 = a cluster to compare this one to (mandatory)
arg #2 = strand option ('strong', 'weak', 'ignore') (not implemented here)
Returns : a TranscriptCluster object, which is empty if the intersection does not contain
any transcript |
Title : length
Usage : $length = $range->length();
Function: get/set the length of this range
Returns : the length of this range
Args : optionaly allows the length to be set
: using $range->length($length) |
new() initializes the attributes:
_transcript_array
_transcriptID_array
_start
_end |
function to include one or more transcripts in the cluster.
Useful when creating a cluster. It takes as argument an array of transcripts, it returns nothing. |
function to set included biotypes in this cluster of transcripts |
Title : start
Usage : $start = $transcript_cluster->end();
Function: get/set the start of the range covered by the cluster. This is re-calculated and set everytime
a new transcript is added to the cluster
Returns : a number
Args : optionaly allows the start to be set |
Title : strand
Usage : $strand = $transcript->strand();
Function: get/set the strand of the transcripts in the cluster.
The strand is set in put_Transcripts when the first transcript is added to the cluster, in that
that method there is also a check for strand consistency everytime a new transcript is added
Returns 0/undef when strand not set
Returns : the strandidness (-1, 0, +1)
Args : optionaly allows the strand to be set |
it returns a string containing the information about the transcripts in the TranscriptCluster object |
Title : union
Usage : $union_cluster = $cluster1->union(@clusters);
Function: returns the union of clusters
Args : a TranscriptCluster or list of TranscriptClusters to find the union of
Returns : the TranscriptCluster object containing all of the ranges |
Methods code
sub _get_start_end
{ my ($self,$t) = @_;
my @exons = sort { $a->start <=> $b->start } @{$t->get_all_Exons};
my $start = $exons[0]->start;
my $end = $exons[-1]->end;
return ($start, $end, $exons[0]->strand);} | sub end
{ my ($self,$end) = @_;
if ($end){
throw( "$end is not an integer") unless $end =~/^[-+]?\d+$/;
$self->{'_end'} = $end;
}
return $self->{'_end'};
}
} | sub exon_Density
{ my ($self, $transcript) = @_;
my $density;
my $exon_span;
my @exons = @{$transcript->get_all_Exons};
@exons = sort { $a->start <=> $b->start } @exons;
my $transcript_length = $exons[$#exons]->end - $exons[0]->start;
foreach my $exon ( @exons ){
$exon_span += $exon->length;
}
$density = $exon_span/$transcript_length;
return $density;} | sub get_ExonCluster
{ my ( $self , $ignore_strand) = @_;
my @clusters;
foreach my $trans (@{$self->get_Transcripts}) {
my $tr_biotype = $trans->biotype;
foreach my $exon (@{$trans->get_all_Exons}) {
my @matching_clusters;
CLUSTER: foreach my $cluster (@clusters) {
if (!($exon->start >= $cluster->end ||
$exon->end <= $cluster->start)) {
if ($cluster->strand eq $exon->strand ){
push (@matching_clusters, $cluster);
}
}
}
if (scalar(@matching_clusters) == 0) {
my $newcluster = Bio::EnsEMBL::Analysis::Tools::Algorithms::ExonCluster->new() ;
$newcluster->add_exon($exon,$trans,$ignore_strand);
push(@clusters,$newcluster);
} elsif (scalar(@matching_clusters) == 1) {
$matching_clusters[0]->add_exon($exon,$trans, $ignore_strand);
} else {
print STDERR "Merging clusters for " . $exon->dbID ."\n";
my @new_clusters;
my $merged_cluster = Bio::EnsEMBL::Analysis::Tools::Algorithms::ExonCluster->new() ;
foreach my $clust (@matching_clusters) {
$merged_cluster->merge($clust, $ignore_strand);
}
$merged_cluster->add_exon($exon,$trans, $ignore_strand);
push @new_clusters,$merged_cluster;
foreach my $clust (@clusters) {
my $found = 0;
MATCHING: foreach my $m_clust (@matching_clusters) {
if ($clust == $m_clust) {
$found = 1;
last MATCHING;
}
}
if (!$found) {
push @new_clusters,$clust;
}
}
@clusters = @new_clusters;
}
}
}
for my $c (@clusters) {
for my $e(@{ $c->get_all_Exons_in_ExonCluster} ) {
$e->cluster($c) ;
}
}
return @clusters;} | sub get_ExonCluster_using_all_Exons
{ my ( $self, $ignore_strand) = @_;
my @clusters;
foreach my $trans (@{$self->get_Transcripts}) {
my $tr_biotype = $trans->biotype;
foreach my $exon (@{$trans->get_all_Exons}) {
my @matching_clusters;
CLUSTER: foreach my $cluster (@clusters) {
$cluster->start. " to " . $cluster->end . " ".$cluster->strand ."\t";
if (!($exon->start >= $cluster->end ||
$exon->end <= $cluster->start)) {
if (!$ignore_strand) {
if ($cluster->strand eq $exon->strand ){
push (@matching_clusters, $cluster);
}
} else {
push (@matching_clusters, $cluster);
}
}
}
if (scalar(@matching_clusters) == 0) {
my $newcluster = Bio::EnsEMBL::Analysis::Tools::Algorithms::ExonCluster->new() ;
$newcluster->add_exon($exon,$trans, $ignore_strand);
push(@clusters,$newcluster);
} elsif (scalar(@matching_clusters) == 1) {
$matching_clusters[0]->add_exon_if_not_present($exon,$trans, $ignore_strand);
} else {
print STDERR "Merging clusters for " . $exon->dbID ."\n";
my @new_clusters;
my $merged_cluster = Bio::EnsEMBL::Analysis::Tools::Algorithms::ExonCluster->new() ;
foreach my $clust (@matching_clusters) {
$merged_cluster->merge_new_exon($clust, $ignore_strand);
}
$merged_cluster->add_exon_if_not_present($exon,$trans, $ignore_strand);
push @new_clusters,$merged_cluster;
foreach my $clust (@clusters) {
my $found = 0;
MATCHING: foreach my $m_clust (@matching_clusters) {
if ($clust == $m_clust) {
$found = 1;
last MATCHING;
}
}
if (!$found) {
push @new_clusters,$clust;
}
}
@clusters = @new_clusters;
}
}
}
return @clusters;} | sub get_IntronClusters
{ my ( $self , $ignore_strand) = @_;
my @clusters;
foreach my $trans (@{$self->get_Transcripts}) {
my @introns = @{get_all_Introns_between_translateable_Exons($trans)};
my $tr_biotype = $trans->biotype;
foreach my $intron (@{get_all_Introns_between_translateable_Exons($trans)}) {
my @matching_clusters;
CLUSTER: foreach my $cluster (@clusters) {
if ($intron->start == $cluster->start && $intron->end == $cluster->end) {
if (!$ignore_strand) {
if ($cluster->strand eq $intron->strand ){
push (@matching_clusters, $cluster);
}
} else {
push (@matching_clusters, $cluster);
}
}
}
if (scalar(@matching_clusters) == 0) {
my $newcluster = Bio::EnsEMBL::Analysis::Tools::Algorithms::IntronCluster->new() ;
$newcluster->{'_types_sets'} = $self->{'_types_sets'};
$newcluster->put_Introns([$intron],$trans);
push(@clusters,$newcluster);
} elsif (scalar(@matching_clusters) == 1) {
$matching_clusters[0]->put_Introns([$intron],$trans);
} else {
throw("Intron cannot match more than one cluster or it will not be unique");
}
}
}
return @clusters;} | sub get_Transcripts
{ my $self = shift @_;
return $self->{'_transcript_array'};
}
} | | get_all_Introns_between_translateable_Exons | description | prev | next | Top |
sub get_all_Introns_between_translateable_Exons
{ my ($transcript) = @_;
my @introns;
my @exons = @{$transcript->get_all_translateable_Exons};
for (my $i = 0; $i < scalar(@exons) - 1; $i++) {
my $intron = new Bio::EnsEMBL::Intron($exons[$i],$exons[$i+1]);
push @introns, $intron;
}
return\@ introns;
}
1;} | sub get_biotypes_included
{ my ($self) = @_ ;
my @bt = keys %{$self->{_biotypes} } ;
return\@ bt ;} | sub get_coding_ExonCluster
{ my ( $self, $ignore_strand ) = @_;
my @clusters;
foreach my $trans (@{$self->get_Transcripts}) {
my $tr_biotype = $trans->biotype;
foreach my $exon (@{$trans->get_all_translateable_Exons}) {
my @matching_clusters;
CLUSTER: foreach my $cluster (@clusters) {
if (!($exon->start >= $cluster->end ||
$exon->end <= $cluster->start)) {
if (!$ignore_strand) {
if ($cluster->strand eq $exon->strand ){
push (@matching_clusters, $cluster);
}
} else {
push (@matching_clusters, $cluster);
}
}
}
if (scalar(@matching_clusters) == 0) {
my $newcluster = Bio::EnsEMBL::Analysis::Tools::Algorithms::ExonCluster->new() ;
$newcluster->add_exon($exon,$trans, $ignore_strand);
push(@clusters,$newcluster);
} elsif (scalar(@matching_clusters) == 1) {
$matching_clusters[0]->add_exon_if_not_present($exon,$trans, $ignore_strand);
} else {
print STDERR "Merging clusters for " . $exon->dbID ."\n";
my @new_clusters;
my $merged_cluster = Bio::EnsEMBL::Analysis::Tools::Algorithms::ExonCluster->new() ;
foreach my $clust (@matching_clusters) {
$merged_cluster->merge_new_exon($clust, $ignore_strand);
}
$merged_cluster->add_exon_if_not_present($exon,$trans, $ignore_strand);
push @new_clusters,$merged_cluster;
foreach my $clust (@clusters) {
my $found = 0;
MATCHING: foreach my $m_clust (@matching_clusters) {
if ($clust == $m_clust) {
$found = 1;
last MATCHING;
}
}
if (!$found) {
push @new_clusters,$clust;
}
}
@clusters = @new_clusters;
}
}
}
return @clusters;} | sub intersection
{ my ($self, $cluster, $ignore_strand) = @_;
if ( scalar( $self->get_Transcripts) == 0 ){
warning( "cluster $self is empty, returning an empty TranscriptCluster");
my $empty_cluster = Bio::EnsEMBL::Analysis::Runnable::Condense_EST::TranscriptCluster->new();
return $empty_cluster;
}
if ( scalar( $cluster->get_Transcripts ) == 0 ){
warning( "cluster $cluster is empty, returning an empty TranscriptCluster");
my $empty_cluster =Bio::EnsEMBL::Analysis::Runnable::Condense_EST::TranscriptCluster->new();
return $empty_cluster;
}
my @transcripts = $self->get_Transcripts;
push( @transcripts, $cluster->get_Transcripts);
my %list;
foreach my $transcript (@transcripts){
$list{$transcript} = $transcript;
}
@transcripts = values( %list );
my ($inter_start,$inter_end);
my ($start1,$end1) = ($self->start , $self->end);
my ($start2,$end2) = ($cluster->start,$cluster->end);
if ( $self->overlaps( $cluster ) ){
$inter_start = ($start2 > $start1) ? $start2 : $start1;
$inter_end = ($end2 < $end1) ? $end2 : $end1;
}
else{
warning( "clusters $self and $cluster do not intersect range-wise, returning an empty TranscriptCluster");
my $empty_cluster =Bio::EnsEMBL::Analysis::Runnable::Condense_EST::TranscriptCluster->new();
return $empty_cluster;
}
my $inter_cluster =Bio::EnsEMBL::Analysis::Runnable::Condense_EST::TranscriptCluster->new() ;
my @inter_transcripts;
foreach my $transcript ( @transcripts ){
my ($start,$end) = $self->_get_start_end($transcript);
if ($start >= $inter_start && $end <= $inter_end ){
$inter_cluster->put_Transcripts( $ignore_strand, $transcript );
}
}
if ( scalar( $inter_cluster->get_Transcripts ) == 0 ){
warning( "cluster $inter_cluster is empty, returning an empty TranscriptCluster");
return $inter_cluster;
}
else{
return $inter_cluster;
}
}
} | sub length
{ my $self = shift @_;
if (@_){
$self->confess( ref($self)."->length() is read-only");
}
return ( $self->{'_end'} - $self->{'_start'} + 1 );
}
} | sub new
{ my ($class,$whatever)=@_;
if (ref($class)){
$class = ref($class);
}
my $self = {};
bless($self,$class);
if ($whatever){
throw( "Can't pass an object to new() method. Use put_Genes() to include Bio::EnsEMBL::Gene in cluster");
}
$self->{_biotype}={} ;
return $self;
}
} | sub put_Transcripts
{ my ($self, $ignore_strand, @new_transcripts)= @_;
throw("Can't add no transcripts to ".$self) if(!@new_transcripts || !$new_transcripts[0]);
if ( !$new_transcripts[0]->isa('Bio::EnsEMBL::Transcript') ){
throw( "Can't accept a [ $new_transcripts[0] ] instead of a Bio::EnsEMBL::Transcript");
}
my $min_start = undef;
my $max_end = undef;
foreach my $transcript (@new_transcripts) {
print "got transcript ".$transcript->stable_id."\n";
my ($start, $end) = $self->_get_start_end($transcript);
if (!defined($min_start) || $start < $min_start) {
$min_start = $start;
}
if (!defined($max_end) || $end > $max_end) {
$max_end = $end;
}
}
foreach my $transcript (@new_transcripts){
if (!$ignore_strand) {
my @exons = @{$transcript->get_all_Exons};
unless ( $self->strand ){
$self->strand( $exons[0]->strand );
}
if ( $self->strand != $exons[0]->strand ){
warning( "You're trying to put $transcript in a cluster of opposite strand");
}
} else {
}
}
unless ( $self->start ){
$self->start( $min_start );
}
unless ( $self->end ){
$self->end( $max_end);
}
if ($min_start < $self->start ){
$self->start( $min_start );
}
if ( $max_end > $self->end ){
$self->end( $max_end );
}
push ( @{ $self->{'_transcript_array'} }, @new_transcripts );
}
} | sub register_biotype
{ my ($self,$biotype) =@_ ;
${$self->{_biotypes} }{$biotype}=() ;} | sub start
{ my ($self,$start) = @_;
if ($start){
throw( "$start is not an integer") unless $start =~/^[-+]?\d+$/;
$self->{'_start'} = $start;
}
return $self->{'_start'};
}
} | sub strand
{ my ($self,$strand) = @_;
if ($strand){
$self->{'_strand'} = $strand;
}
return $self->{'_strand'};
}
} | sub to_String
{ my $self = shift @_;
my $data='';
foreach my $tran ( @{ $self->{'_transcript_array'} } ){
my @exons = @{$tran->get_all_Exons};
my $id;
if ( $tran->stable_id ){
$id = $tran->stable_id;
}
else{
$id = $tran->dbID;
}
$data .= sprintf "Id: %-16s" , $id;
$data .= sprintf "Contig: %-21s" , $exons[0]->contig->id;
$data .= sprintf "Exons: %-3d" , scalar(@exons);
my ($start, $end) = $self->_get_start_end($tran);
$data .= sprintf "Start: %-9d" , $start;
$data .= sprintf "End: %-9d" , $end;
$data .= sprintf "Strand: %-3d" , $exons[0]->strand;
$data .= sprintf "Exon-density: %3.2f\n", $self->exon_Density($tran);
}
return $data;
}
} | sub union
{my ($self,@clusters, $ignore_strand) = @_;
if ( ref($self) ){
unshift @clusters, $self;
}
my $union_cluster = Bio::EnsEMBL::Analysis::Runnable::Condense_EST::TranscriptCluster->new() ;
my $union_strand;
foreach my $cluster (@clusters){
unless ($union_strand){
$union_strand = $cluster->strand;
}
unless ( $cluster->strand == $union_strand){
warning("You're making the union of clusters in opposite strands");
}
$union_cluster->put_Transcripts($ignore_strand, $cluster->get_Transcripts);
}
return $union_cluster;
}
} | General documentation
Methods start and end are typical for a range. We also implement the boolean
and geometrical methods for a range.
These methods return true or false. They throw an error if start and end are
not defined. They are implemented in Bio::RangeI.
$cluster->overlaps($other_cluster) && print "Clusters overlap\n";
Title : overlaps
Usage : if($cluster1->overlaps($cluster)) { do stuff }
Function: tests if $cluster2 overlaps $cluster1 overlaps in the sense of genomic-range overlap,
it does NOT test for exon overlap.
Args : arg #1 = a range to compare this one to (mandatory)
arg #2 = strand option ('strong', 'weak', 'ignore')
Returns : true if the clusters overlap, false otherwise
Title : contains
Usage : if($cluster1->contains($cluster2) { do stuff }
Function: tests whether $cluster1 totally contains $cluster2
Args : arg #1 = a range to compare this one to (mandatory)
alternatively, integer scalar to test
arg #2 = strand option ('strong', 'weak', 'ignore') (optional)
Returns : true if the argument is totaly contained within this range
Title : equals
Usage : if($cluster1->equals($cluster2))
Function: test whether the range covered by $cluster1 has the same start, end, length as the range
covered by $cluster2
Args : a range to test for equality
Returns : true if they are describing the same range
These methods do things to the geometry of ranges, and return
Bio::RangeI compliant objects or triplets (start, stop, strand) from
which new ranges could be built. They are implemented here and not in Bio::RangeI, since we
want them to return a new TranscriptCluster object.
Title : overlap_extent
Usage : ($a_unique,$common,$b_unique) = $a->overlap_extent($b)
Function: Provides actual amount of overlap between two different
ranges. Implemented already in RangeI
Example :
Returns : array of values for
- the amount unique to a
- the amount common to both
- the amount unique to b
Args :