Re: [Treebase-devel] viz problem with trees for S11267

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 422-6466

Hi folks,

Just joined the list, as I am interested in using TB to develop
collaborative methods for synthesizing plant phylogeny, e.g., by grafting
clades together and other kinds of agglomeration.  So, I am particularly
interested in the API and harvesting, especially with respect to taxonomic
names and classifications, and GenBank identifiers of sequences.

Speaking of which, how would I go about harvesting the GenBank numbers for a
given study and associating them with their alignments?  Its that possible
with the current API?

I like Rutger's plug-in ideas.  In general, it's easier for someone like me
to provide a simple web service, rather than contribute directly to TB
development.

-Rick

On Thu, Apr 14, 2011 at 7:24 AM, Rutger Vos <R....@re...> wrote:

> I've been thinking about a redesign lately, and here's what I would do:
>
> - make sure we can export all the metadata from TreeBASE in (nexml)
> files, i.e. Laurel's GSoC project
>
> - get all the files for all the studies and create a simple folder
> structure with those files, more or less along the lines of the
> phylows urls (e.g phylows/tree/TB2/nexml/T1312), in all file formats
> we can think of (json, nexus, phylip, newick, phyloxml, fasta, static
> html, etc...)
>
> - create some mod_rewrite rules to map our urls onto the folder
> structure (so going from phylows/tree/TB2:T1312?format=nexml to the
> actual file). Performance will obviously be much, much better.
>
> - also allow harvesting of those files using rsync and ftp. Now we
> have data dumps.
>
> - create a simple plug in architecture where, given a query string, a
> remote web service returns a list of hits which it somehow generates
> from its local, harvested data dump
>
> Here's a use case: Laurel wants to implement BLASTing into TreeBASE.
> So she periodically harvests everything in phylows/matrix/TB2/fasta
> with an rsync cron job. She runs formatdb on the fasta sequences,
> creating a standalone BLAST. She then writes a simple cgi script that
> accepts a target string (e.g.
> myblast.cgi?tb2.blast=acgctcgcatcgcatcgactacgac) and returns a list of
> phylows matrix urls from the matching results.
>
> On the TreeBASE side, we simply add a search widget that delegates the
> query to Laurel's service and integrates it into our interfaces
> (graphical, web services).
>
> With an architecture like that it would be so much easier for anyone
> to add functionality in whatever programming language they like
> without having to deal with a massive database schema. Of course any
> of those remote services might have its own little database, but it'd
> be more along the lines of a three-table SQLite database to store the
> ITIS taxonomy structure such that we can find all TreeBASE taxa
> subtended by a given ITIS higher taxon ID (for example).
>
> We'd have to implement some core plug ins ourselves, notably one that
> extracts the metadata (using RDFA2RDFXML) and sticks that in a triple
> store so we can search on, say, author names, journals, etc. I think
> that's scalable because it's only a few dozen triples for each study.
>
> The idea is a little bit inspired by DAS, which seems to work quite
> well: http://www.biodas.org/wiki/Main_Page
>
> (I'm leaving out the submission part as an exercise for the reader.
> Presumably there would have to be a restricted area where properly
> formatted files are uploaded and made available to reviewers.)
>
> Rutger
>
> p.s. I've done some harvesting a few weeks back, but unless that's
> created queries that are still running I'm innocent.
>
> On Thu, Apr 14, 2011 at 12:28 PM, Roderic Page <r....@bi...>
> wrote:
> > TreeBASE  performance has nothing to do with me folks, I pretty much gave
> up trying to download data from it a few weeks back. Someone really, really
> needs to rethink the way TreeBASE works, because it's virtually unusable.
> >
> > Regards
> >
> > Rod
> >
> > On 14 Apr 2011, at 04:20, William Piel wrote:
> >
> >>
> >> On Apr 13, 2011, at 9:36 PM, Hilmar Lapp wrote:
> >>
> >>> The trees in S11267 seem to silently fail to render in PhyloWidget.
> >>> All I get is a single dot. At least try the first three:
> >>>
> >>> http://www.treebase.org/treebase-web/search/study/trees.html?id=11267
> >>>
> >>> Is this a temporary glitch, a problem with the reconstructed file, or
> >>> something else? Should I best file this as a bug in the bug tracker?
> >>>
> >>>      -hilmar
> >>>
> >>
> >> Thanks for the alert. This is actually a bug in PhyloWidget -- if the
> word "tree" appears in the title of the tree block, PhyloWidget confuses
> this word with the TREE command, and so goofs up the parsing.
> >>
> >> I have removed the word "tree" from the tree block, so it works now.
> >>
> >> However, you might need to refresh the PhyloWidget window after loading
> to get the tree to pull through -- TreeBASE feels quite slow lately; our API
> must be getting hit by  someone (Rod?).
> >>
> >> bp
> >>
> >>
> >>
> >>
> ------------------------------------------------------------------------------
> >> Benefiting from Server Virtualization: Beyond Initial Workload
> >> Consolidation -- Increasing the use of server virtualization is a top
> >> priority.Virtualization can reduce costs, simplify management, and
> improve
> >> application availability and disaster protection. Learn more about
> boosting
> >> the value of server virtualization.
> http://p.sf.net/sfu/vmware-sfdev2dev
> >> _______________________________________________
> >> Treebase-devel mailing list
> >> Tre...@li...
> >> https://lists.sourceforge.net/lists/listinfo/treebase-devel
> >>
> >
> > ---------------------------------------------------------
> > Roderic Page
> > Professor of Taxonomy
> > Institute of Biodiversity, Animal Health and Comparative Medicine
> > College of Medical, Veterinary and Life Sciences
> > Graham Kerr Building
> > University of Glasgow
> > Glasgow G12 8QQ, UK
> >
> > Email: r....@bi...
> > Tel: +44 141 330 4778
> > Fax: +44 141 330 2792
> > AIM: rod...@ai...
> > Facebook: http://www.facebook.com/profile.php?id=1112517192
> > Twitter: http://twitter.com/rdmpage
> > Blog: http://iphylo.blogspot.com
> > Home page: http://taxonomy.zoology.gla.ac.uk/rod/rod.html
> >
> >
> >
> >
> >
> >
> >
> >
> >
> ------------------------------------------------------------------------------
> > Benefiting from Server Virtualization: Beyond Initial Workload
> > Consolidation -- Increasing the use of server virtualization is a top
> > priority.Virtualization can reduce costs, simplify management, and
> improve
> > application availability and disaster protection. Learn more about
> boosting
> > the value of server virtualization. http://p.sf.net/sfu/vmware-sfdev2dev
> > _______________________________________________
> > Treebase-devel mailing list
> > Tre...@li...
> > https://lists.sourceforge.net/lists/listinfo/treebase-devel
> >
>
>
>
> --
> Dr. Rutger A. Vos
> School of Biological Sciences
> Philip Lyle Building, Level 4
> University of Reading
> Reading
> RG6 6BX
> United Kingdom
> Tel: +44 (0) 118 378 7535
> http://www.nexml.org
> http://rutgervos.blogspot.com
>
>
> ------------------------------------------------------------------------------
> Benefiting from Server Virtualization: Beyond Initial Workload
> Consolidation -- Increasing the use of server virtualization is a top
> priority.Virtualization can reduce costs, simplify management, and improve
> application availability and disaster protection. Learn more about boosting
> the value of server virtualization. http://p.sf.net/sfu/vmware-sfdev2dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel
>