[Bioperl-l] Re: [BioSQL-l] seqfeature.display_id

Lincoln Stein lstein@cshl.org
Wed, 2 Oct 2002 08:44:52 -0400


I am very happy with the idea of display_name().  The intent is much clearer.   
I second the idea of replacing seqname() with seq_id() or seq_accession().

Are their guidelines on what the following terms (which are used throughout 
the sequence modules) mean?

	name
	accession
	id

My interpretation is:

	name -- human readable, non-unique
	accession -- public identifier, unique across a namespace
	id -- unique internal ID, not for public use

Lincoln

On Tuesday 01 October 2002 01:11 pm, Hilmar Lapp wrote:
> I agree with this. That's why I proposed to
>
> 1) make the meaning absolutely clear and rename display_id to
> display_name (but keep an undocumented alias)
>
> 2) fix the implementation in SeqFeature::Generic to refer to its own
> attribute instead of being an alias for seqname()
>
> Can I take your responses as agreement with me making these changes?
>
> Jason/Heikki, any comment or feeling on this?
>
> 	-hilmar
>
> On Tuesday, October 1, 2002, at 05:25 AM, Ewan Birney wrote:
> > On Tue, 1 Oct 2002, Lincoln Stein wrote:
> >> Was there any further discussion on this?  I do use display_id()
> >> in the
> >> Bio::Graphics module, and am confused about the distinction between
> >> display_id() and seqname().
> >
> > IMNSHO
> >
> > display_id() == thing to be displayed for this object to a user
> >
> > seqname() == name of sequence attached - possibly better called
> > seq_id or
> > seq_accession_number() or something more obviously "real id like".

-- 
Lincoln Stein
lstein@cshl.org