Hi Matthias, If you look at the HTML composite datatype, there is a master file (HTML), and an arbitrary number of arbitrarily named child files like images. The BLAST database datatype follows this, but we have used a simple text file as the master file (just the stdout from makeblastdb) in order to have something for Galaxy to show in the main pane when clicking on these datasets. I can't think of any other composite examples which would be more relevant, hopefully someone else on the list can? Regards, Peter On Fri, Aug 17, 2018 at 2:55 PM, Matthias Bernt <m.bernt@ufz.de> wrote:
Hi,
thanks for your support. This helps.
I have thought a bit about composite data types. And have additional questions.
In my case the additional data is essentially a folder (w subfolders). The contents of the folder vary (it depends on the input of the programs that generate them). So the question is if I can/should
- add a single folder (with all the hirarchy as a single composite data element) .. I have only seen a add_composite_file function.
- add a dynamic number of composite files
- or if it is better to use the extra_files_path mechanism (which seems to be the simplest for me)
In any of these cases my next question would be how to create a test using such a data as input and output (for output I have seen an example somewhere).
Cheers, Matthias
On 17.08.2018 11:53, Peter Cock wrote:
This is probably a John Chilton question, as the Planemo lead.
The way I do it is to "manually" install the datatype into a Galaxy test instance (adding entries to the datatypes_conf.xml and Python files to Galaxy's internal library), and then call ``planemo test`` pointing at this test instance. You can see that approach in action here:
https://github.com/peterjc/galaxy_blast/blob/579c348ced72d4e6f1ef7fb0cded98e...
https://github.com/peterjc/galaxy_blast/blob/579c348ced72d4e6f1ef7fb0cded98e...
and:
https://github.com/peterjc/galaxy_mira/blob/206259620376b322fc8ed99a6efdd371...
https://github.com/peterjc/galaxy_mira/blob/206259620376b322fc8ed99a6efdd371...
There may be a more elegant solution nowadays using planemo to do some of the work.
Peter
On Thu, Aug 16, 2018 at 9:30 PM, Matthias Bernt <m.bernt@ufz.de> wrote:
Dear Peter,
you are right. I hope that this will be much less in the end (I'm still learning about the package).
The main question still remains, how do I get `planemo test` to include the data types defined in the xml file?
Best, Matthias
Am 16/08/18 18:32 schrieb Peter Cock <p.j.a.cock@googlemail.com>:
Defining 20 different text-based formats does not look ideal (if that is what you are doing).
Do you have sample output in the repository? Perhaps at least some of these can be better defined as tabular instead?
Or, perhaps you can define one composite datatype for the folder of output instead?
Peter
On Thu, Aug 16, 2018 at 4:18 PM, Matthias Bernt <m.bernt@ufz.de> wrote:
Hi Peter,
I hope that subclassing simple data types will be sufficient.
More details:
I'm currently trying to (auto)wrap the checkm suite https://github.com/Ecogenomics/CheckM. Current state here:
https://github.com/bernt-matthias/mb-galaxy-tools/tree/master/tools/checkm.
These tools often create an output folder which is then the input to other tools. So I thought to create a (output) data type for each of the tools and save the folder using extra_files_path. Then I can pass the folder to downstream tools (and can check for proper input). I hope that subclassing txt or tabular will be sufficient.
Cheers, Matthias
On 16.08.2018 17:02, Peter Cock wrote:
More details might help - are you just defining the new datatype as a subclass in the XML, or do you need to include Python code (e.g. for a sniffer)?
If you want to see some examples of datatypes using Python code which are available via the Tool Shed, here are two:
https://github.com/peterjc/galaxy_blast/tree/master/data_managers/ncbi_blast... (now also in the Galaxy core)
https://github.com/peterjc/galaxy_mira/tree/master/datatypes/mira_datatypes
Peter
On Thu, Aug 16, 2018 at 3:48 PM, Matthias Bernt <m.bernt@ufz.de> wrote:
Dear list,
just a request for links to documentation: How can I realize tool specific data types. I'm just developing a set of tools that need their own data types, but I don't want to add them to Galaxy's core data types (yet).
I've seen examples of tools that had a datatypes_conf.xml. So I created one, but it seems that it is ignored.
Cheers, Matthias
--
------------------------------------------- Matthias Bernt Bioinformatics Service Molekulare Systembiologie (MOLSYB) Helmholtz-Zentrum für Umweltforschung GmbH - UFZ/ Helmholtz Centre for Environmental Research GmbH - UFZ Permoserstraße 15, 04318 Leipzig, Germany Phone +49 341 235 482296, m.bernt@ufz.de, www.ufz.de
Sitz der Gesellschaft/Registered Office: Leipzig Registergericht/Registration Office: Amtsgericht Leipzig Handelsregister Nr./Trade Register Nr.: B 4703 Vorsitzender des Aufsichtsrats/Chairman of the Supervisory Board: MinDirig Wilfried Kraus Wissenschaftlicher Geschäftsführer/Scientific Managing Director: Prof. Dr. Dr. h.c. Georg Teutsch Administrative Geschäftsführerin/ Administrative Managing Director: Prof. Dr. Heike Graßmann ------------------------------------------- ___________________________________________________________ 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: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/
--
------------------------------------------- Matthias Bernt Bioinformatics Service Molekulare Systembiologie (MOLSYB) Helmholtz-Zentrum für Umweltforschung GmbH - UFZ/ Helmholtz Centre for Environmental Research GmbH - UFZ Permoserstraße 15, 04318 Leipzig, Germany Phone +49 341 235 482296, m.bernt@ufz.de, www.ufz.de
Sitz der Gesellschaft/Registered Office: Leipzig Registergericht/Registration Office: Amtsgericht Leipzig Handelsregister Nr./Trade Register Nr.: B 4703 Vorsitzender des Aufsichtsrats/Chairman of the Supervisory Board: MinDirig Wilfried Kraus Wissenschaftlicher Geschäftsführer/Scientific Managing Director: Prof. Dr. Dr. h.c. Georg Teutsch Administrative Geschäftsführerin/ Administrative Managing Director: Prof. Dr. Heike Graßmann -------------------------------------------
--
------------------------------------------- Matthias Bernt Bioinformatics Service Molekulare Systembiologie (MOLSYB) Helmholtz-Zentrum für Umweltforschung GmbH - UFZ/ Helmholtz Centre for Environmental Research GmbH - UFZ Permoserstraße 15, 04318 Leipzig, Germany Phone +49 341 235 482296, m.bernt@ufz.de, www.ufz.de
Sitz der Gesellschaft/Registered Office: Leipzig Registergericht/Registration Office: Amtsgericht Leipzig Handelsregister Nr./Trade Register Nr.: B 4703 Vorsitzender des Aufsichtsrats/Chairman of the Supervisory Board: MinDirig Wilfried Kraus Wissenschaftlicher Geschäftsführer/Scientific Managing Director: Prof. Dr. Dr. h.c. Georg Teutsch Administrative Geschäftsführerin/ Administrative Managing Director: Prof. Dr. Heike Graßmann -------------------------------------------