[Bioperl-l] Withdraw Bio::Graphics and Bio::DB::SeqFeature from bioperl distribution?

Chris Fields cjfields at illinois.edu
Mon Nov 10 22:58:44 EST 2008


On Nov 10, 2008, at 4:57 PM, Lincoln Stein wrote:

> Hi Folks,
>
> Sorry, I sent my letter out while multitasking, which was  
> inappropriate for such a nuanced subject.
>
> Here's my feeling about the issue, which is very similar to Chris's  
> approach: Bioperl should be split up into a core API containing  
> Bio::Root, the interface modules, and some of the really basic  
> modules such as Bio::SeqIO. There are then a series of separately  
> maintained and released modules that build on top of BioPerl and  
> using it as a dependency. I am not in favor of having a "separate  
> monolithic distribution" however. I'd prefer it to be more like this:
>
> 	• Bio::Perl -- the core distribution, containing Bio::Root,  
> Bio::Seq, BioSeqFeature, Bio::SeqIO and Bio::Annotation
> 	• Bio::Align -- alignment support
> 	• Bio::Ontology -- ontology support
> 	• Bio::Microarray -- microarray support
> 	• Bio::PopGen -- population genetics
> 	• Bio::SeqEvolution -- evolutionary biology
> 	• Bio::Structure -- structures
> 	• Bio::Tree -- trees
> I wonder how many interdependencies there are to disentangle?
>
> In the immediate future, it'd be great to get the regression tests  
> working 100% and do a CPAN release, but this may be easier said than  
> done.
>
> Lincoln

BTW, I would like to add to that list Bio::Tools (Bio::Tools* related  
changes suggested recently by Heikki) and Bio::Dev (for in development  
or untested modules, or experimental modules with an unstable API).

I'll work on a few bugs towards getting 1.6 released with a rough  
timeline for end of Nov, maybe even Thanksgiving.  We can work on  
splitting things up after that.

-c


More information about the Bioperl-l mailing list