Hi Greg!

> > Hi all!
> > 
> >> I'd like to define 'blastxml' to 'tabular' conversion within Galaxy
> >> so that this can be used automatically (e.g. Tools expecting just
> >> tabular data would have converted BLAST XML files listed in the
> >> drop down). The functionality to do this exists in my Python tool
> >> 'blastxml_to_tabular' currently in the 'ncbi_blast_plus' repository,
> >> which might have to move into 'blast_datatypes' for this to work.
> >> http://toolshed.g2.bx.psu.edu/view/devteam/ncbi_blast_plus
> >> http://toolshed.g2.bx.psu.edu/view/devteam/blast_datatypes
> >> 
> >> I've skimmed over the instructions here, but would like to look
> >> at a couple of real examples to make sure I've understood:
> >> http://lists.bx.psu.edu/pipermail/galaxy-dev/2013-February/013434.html
> >> 
> >> Are there any good examples of 3rd party Galaxy datatypes
> >> defining conversion like this?
> > 
> > To add some more question to that topic.
> > How to hide the converter from the tool-panel, when its available within
> > Galaxy?
> This is a known issues, and there is a Trello card for it here.  I'll be 
> working on this issues this week and the fix should be included in the next 
> Galaxy release.
> 
> https://trello.com/card/toolshed-fix-the-process-that-currently-adds-entries-to-the-galaxy-tool-panel/506338ce32ae458f6d15e4b3/642

Great!

> > And is it possible to use third-party tool-shed dependencies in
> > data-type definition files? I have a data-type that is really hard to
> > sniff. I would like to use an external library for that purpose.  

> Including tool dependency definitions inside datatype_conf.xml files is not 
> supported. 
> Can you clarify this a bit?  It would be good to get some details of the 
> specific components you're dealing with so I can get a clear picture of what 
> may be required here.

I'm developing datatypes for chemical structures [1], one of these
datatypes is called SMILES (smi) and its mainly a string that defines
atoms, like "CC(=O)OC1=CC=CC=C1C(=O)O" for aspirin. Its hard to sniff
that datatype. The only good solution I can image is to try to read such
a structure and if I succeed and can create a molecule out of a random
string, I assume its a valid SMILES string. Do that for the first 10
lines or so should be sufficient to define the datatype as 'smi'.
OpenBabel is a library that can read such string and convert them, there
are also python bindings. All of my converters are using OpenBabel, so I
included it as repository_dependency. The problem is now, that if I want
to use the python bindings from openbabel for the datatype definition
file (molecules.py) its not available, because the path is not set. At
least thats what I assume. Because if I install OpenBabel globally it
works just fine.

To the original question from Peter, is there a way to integrate
converters within galaxy? My converters only appear in the tool-pane?

Thanks and a happy weekend!
Bjoern

> > 
> > Thanks,
> > Bjoern
> > 
> > 
> 
> 
> Thanks!
> 
> Greg Von Kuster
> 
> 
> > ___________________________________________________________
> > Please keep all replies on the list by using "reply all"
> > in your mail client.  To manage your subscriptions to this
> > and other Galaxy lists, please use the interface at:
> > 
> > http://lists.bx.psu.edu/
> 


___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

  http://lists.bx.psu.edu/

Reply via email to