search

bioperl / bioperl-live-redmine

Legacy tickets migrated from the OBF Redmine issue tracker: http://redmine.open-bio.org
0 stars 0 forks source link

Bio::Assembly::Scaffold Source #29

Open cjfields opened 9 years ago

cjfields commented 9 years ago

Author Name: Florent Angly (@fangly) Original Redmine Issue: 2370, https://redmine.open-bio.org/issues/2370 Original Date: 2007-09-21 Original Assignee: Bioperl Guts

CVS version…

The module Bio::Assembly::Scaffold pod mentions that the method new can have the attribute -source to specify the assembly program used to generate the assembly. In fact, this attribute is not implemented.

I see 2 cases: i/ -source is estimated to be unnecessary and there should be no mention of it in the pod ii/ -source is estimated to be useful and should be implemented. In that case, it should be taken into acount that Bio::Assembly::Contig also have a source attribute. Probably contig objects should have by default the source defined for their assembly object. If contigs are added to the assembly and have a different source that that of the assembly, the source of the assembly should be changed to something like ‘Various’, or something like that.

So which is it?

cjfields commented 9 years ago

Original Redmine Comment Author Name: Chris Fields Original Date: 2007-09-24T13:20:32Z

(In reply to comment #0)

CVS version…

The module Bio::Assembly::Scaffold pod mentions that the method new can have the attribute -source to specify the assembly program used to generate the assembly. In fact, this attribute is not implemented.

I see 2 cases: i/ -source is estimated to be unnecessary and there should be no mention of it in the pod ii/ -source is estimated to be useful and should be implemented. In that case, it should be taken into acount that Bio::Assembly::Contig also have a source attribute. Probably contig objects should have by default the source defined for their assembly object. If contigs are added to the assembly and have a different source that that of the assembly, the source of the assembly should be changed to something like ‘Various’, or something like that.

So which is it?

I think it would be easier to just remove any mention of it. As you state the source for each Contig could theoretically be different, so implementing Scaffold::source() might be more ambiguous/confusing. If you agree I’ll go ahead and remove it from POD.

As I’ve indicated several times in the past, I believe it’s worth thinking about reimplementing parts of Bio::Assembly, or rethinking how we want Assembly to handle seq assembly data in a lightweight manner. For instance, I think the general set of Bio::Assembly classes and interfaces works but there are several significant issues with Bio::Align::AlignI interface, use of tied DB handles via SeqFeature::Collection, and a number of unimplemented methods which need to be addressed. Most of those might be fixed within the IO part of Bio::Assembly.

cjfields commented 9 years ago

Original Redmine Comment Author Name: Florent Angly Original Date: 2007-09-24T14:54:28Z

Chris, I have a preference for ii/ because why should only contigs have source and why not scaffolds and singlets as well? But really, any way is fine with for me, so please go ahead and update the POD.

cjfields commented 9 years ago

Original Redmine Comment Author Name: Chris Fields Original Date: 2007-09-24T15:07:31Z

(In reply to comment #2)

Chris, I have a preference for ii/ because why should only contigs have source and why not scaffolds and singlets as well? But really, any way is fine with for me, so please go ahead and update the POD.

I can see implementing Singlet::source() for consistency. We can implement source() in Scaffold as well but not to be explicitly set (which is where I think it would be too ambiguous). Scaffold::source() would just return the source() for the various contig/singlets if they are the same or ‘various’ if more than one source program is used for the contained objects.

I’m changing this to an enhancement request as it doesn’t impact anything directly. May be time soon to get a bug stomping session going.

cjfields commented 9 years ago

Original Redmine Comment Author Name: Chris Fields Original Date: 2008-02-12T10:53:08Z

Florent, any progress on Bio::Assembly-related issues?

cjfields commented 9 years ago

Original Redmine Comment Author Name: Chris Fields Original Date: 2008-04-08T21:46:37Z

Pushing to 1.7.

cjfields commented 9 years ago

Original Redmine Comment Author Name: Chris Fields Original Date: 2011-04-05T21:30:12Z

Moving to Feature tracker based on comments.