[Bioperl-l] [Gmod-gbrowse] FeatureIO refactoring

Scott Cain scott at scottcain.net
Tue Feb 2 17:06:07 UTC 2010


Hi Chris,

This sounds like a really good approach; I particularly like (2) and
tying FeatureIO to SeqFeature::Store.

About ontology checking: I think the only thing that ought to be
(optionally) checked is the type (column 3) against SO (not SOFA).  It
is true that SeqFeature::Store does no type checking, and I don't know
that it needs to, but having the option of doing that through
FeatureIO would be a good thing.

Thanks for getting this started,
Scott


On Tue, Feb 2, 2010 at 9:04 AM, Chris Fields <cjfields at illinois.edu> wrote:
> All,
>
> Based on a few discussions from the BioPerl meeting, I've decided to go ahead, bite the bullet, and (on a branch) pull out the Bio::FeatureIO code from core for refactoring.  FeatureIO-based code now lies in the Bio-FeatureIO repository in bioperl's SVN.  The resulting code (FeatureIO-less) core probably won't be merged back into main trunk until after a 1.6.2 release, at which point we may just proceed towards the long-awaited restructuring scheme for BioPerl.
>
> I'm pretty open for discussion on how to proceed.  My current somewhat loose plans:
>
> 1) Rewrite FeatureIO code to remove dependencies specifically on Bio::SeqFeature::Annotated (BSFA).
>
> There have been several arguments on and off list as to BSFA's issues.  I think, due to BSFA's idiosyncracies, it's best to remove use of it from BioPerl and replace it with something more lightweight.
>
> 2) Consolidate FeatureIO around Bio::DB::SeqFeature::Store's Loader classes and built-in SeqFeature methods for generating GFF data.
>
> No need to rehash code if it's already working well (and an opportunity to fix code that needs it).  In particular, the GFF3 loader appears to be working very well.  However, if there are any caveats, this would be the time to air them!  Also, with that mind...
>
> 3) Optional type-checking against SO/SOFA using tools already in place (Bio::Ontology::*)
>
> This is a bit trickier, as I don't recall Bio::DB::SeqFeature having any typing mechanism, I'm guessing in keeping with it's lightweight nature.  Should we use the attribute Ontology_term for this?
>
> 4) Of course, addressing any FeatureIO-related bugs present in Bugzilla.
>
>
>
> Feedback and suggestions welcome!
>
> chris
> ------------------------------------------------------------------------------
> The Planet: dedicated and managed hosting, cloud storage, colocation
> Stay online with enterprise data centers and the best network in the business
> Choose flexible plans and management services without long-term contracts
> Personal 24x7 support from experience hosting pros just a phone call away.
> http://p.sf.net/sfu/theplanet-com
> _______________________________________________
> Gmod-gbrowse mailing list
> Gmod-gbrowse at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
>



-- 
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research




More information about the Bioperl-l mailing list