[Bioperl-l] new/unannounced methods

Chris Fields cjfields at uiuc.edu
Sat Mar 10 15:47:32 EST 2007

On Mar 10, 2007, at 11:52 AM, Hilmar Lapp wrote:

> On Mar 9, 2007, at 9:55 PM, Chris Fields wrote:
>>> Is anyone out there currently working on this? If not I am
>>> volunteering.
>>> I'd like to move Bio::SeqUtils::trunc_with_features() and friends to
>>> Bio::SeqI::trunc(), etc.
>> I don't think you'll hear too many protests if you want to take it
>> on.  Just make sure to add decent tests as needed and document
>> everything (one lesson learned from all this!).
>> The trunc() method is implemented in Bio::PrimarySeqI, so it probably
>> should stay there.
> Yes, but PrimarySeqI's don't have features.
> You could say that it may nevertheless know that there is a derived
> class of Seqs that do. However, why not override the method in, e.g.,
> Bio::SeqI, call the inherited function, and then add whatever is
> needed to truncate the features, too.
> 	-hilmar

Yes, that's true; it may make it simpler to keep that code confined  
to SeqI.  My point was that PrimarySeqI::trunc() already seems to  
anticipate the object might be something else besides a simple  
PrimarySeq so one could use isa->('FeatureHolderI') there:

    my $seqclass;
    if($self->can_call_new()) {
        $seqclass = ref($self);
    } else {
        $seqclass = 'Bio::PrimarySeq';

    my $out = $seqclass->new( '-seq' => $str,
			     '-display_id'  => $self->display_id,
			     '-accession_number' => $self->accession_number,
			     '-alphabet' => $self->alphabet,
			     '-desc' => $self->desc(),
                              '-verbose' => $self->verbose
    return $out;

but it would be just as easy to do this in Bio::SeqI after calling  


More information about the Bioperl-l mailing list