[BioPython] biopython in ZOPE
Jeffrey Chang
jchang@smi.stanford.edu
Thu, 28 Mar 2002 20:55:54 -0800
Yes, there's definitely interest here in wrapping biopython code in
Zope objects. My feeling is that the best way to do this is to build
a Zope product that accesses biopython. That way, you can just
distribute the product and let people embed biopython functionality
into Zope-based websites. However, I'm really just a Zope newbie, so
you could probably come up with better solutions than I.
I believe Johann Visagie came up with the idea maybe two years ago.
However, there's been a dearth of people who have both the skills and
the time to put it together!
Jeff
On Thu, Mar 28, 2002 at 12:15:52PM +0100, Danny Navarro wrote:
> Hi all,
>
> I am a "curious" biochemistry/bioinformatics undergraduate student who
> always have liked very much computers and specially Open Source.
>
> Right now, I am able to write some code in Python and Perl, I have some
> skills administrating Debian servers,... and I know more "just for fun"
> computer things.
>
> But I don't know what can I do to be useful for the open bioinformatics
> community. I would like to help to make it stronger.
>
> One thing I am very enthusiastic about is the idea of wrapping biopython
> modules in ZOPE, so that you can construct web sites with ZOPE and at
> the same time provide bioinformatics web services using biopython
> modules.
>
> What it'll be really cool woud be the intercommunication between
> biopython and ZOPE APIs. To do this:
>
> would be better to extend ZOPE adding a "biopython product" using
> Zclasses?
>
> ...or would be better the implementation in the biopython API the
> accession to the ZOPE one? (like bioperl with MySQL)
>
> ... or both might be required?
>
> I have seen some recent messages in the mailing list archive regarding
> to this matter:
>
> http://www.biopython.org/pipermail/biopython-dev/2002-January/000802.html
>
> ==================================================
> > Biopython zope objects complete with interface (Blast, .....) and
> output
> > (BioXML,....)
> > Bioperl zope objects complete with interface (T-coffee) and output
> > (BioXML, GAME, ...)
> > Web maintainers could just install the products and Voila.
> > Biopython ZClasses (and maybe BioPerl objects?) could be
> subclassed for
> > special uses through properties interface.
> > The output of one zope product could be fed into another allowing
> for
> > complex scripts.
> > A weird example.
> > From Swissprot chose trypsin->fasta->phylogeny->align each
> > group->consesus for each group->common restriction enzymes, .....
>
> I think these are the sort of ideas that need to be shared with the
> Biopython (and possibly Bioperl) developers! :-)
> ===================================================
>
> http://www.biopython.org/pipermail/biopython/2000-August/000350.html
>
> ===================================================
> And here's the kicker:
>
> - Existing Python modules can be imported into Zope's namespace and
> be
> exposed to the Web. With a little bit of wrapping, it may
> eventually be
> possible to build a set of reusable "web widgets" to wrap
> Biopython
> objects. This is still a very vaguely formed idea, but it is
> where I think
> I would like to work towards, eventually. It would mean that the
> site
> could actually contribute towards the Biopython project in a more
> concrete
> manner than just providing PR.
>
> Now firstly... do we have the go-ahead to start spending some time
> on this?
> Jeff? :-)
> =====================================================
>
> Ok, so these kinds things I would really like to do.
>
> So if you want a quick learner volunteer to colaborate in this possible
> project just say me what I can do, please.
>
> Danny
>
> _______________________________________________
> BioPython mailing list - BioPython@biopython.org
> http://biopython.org/mailman/listinfo/biopython