Skip to content

[Bioc-devel] Package Classifications via BiocViews

6 messages · Shepherd, Lori, Hervé Pagès, Burak Ogan Mancarci

Original
Burak Ogan Mancarci
# 
Hi,

I am a developer working on the bioconductor package gemma.R
<https://bioconductor.org/packages/3.16/data/experiment/html/gemma.R.html>.
On publication we have used the ExperimentData biocview to classify the
package but we believe this was not the correct decision since the
package's function is to access a database much like geoquery. I was
wondering if it was safe to change the classification on our end by
changing biocviews directly or if it would cause issues with the
bioconductor infrastructure?

Cheers,
Ogan
Shepherd, Lori
# 
Changing the biocviews is not sufficient.  There needs to be work on our end to change the manifest the package belongs to and to clear the build products from the data repositories so it is only found in one location.

Are you sure you want to proceed with switching?  It is not trivial on our end.


Lori Shepherd - Kern

Bioconductor Core Team

Roswell Park Comprehensive Cancer Center

Department of Biostatistics & Bioinformatics

Elm & Carlton Streets

Buffalo, New York 14263
Burak Ogan Mancarci
# 
Apologies for the inconvenience.

The change isn't essential on our end and can be avoided or delayed if you
think the ExperimentData label can be applied to a package responsible for
accessing data from a database rather than storing static data. We were
only worried about misleading use of biocviews terms.

Cheers,
Ogan

On Wed., Sep. 28, 2022, 14:26 Kern, Lori, <Lori.Shepherd at roswellpark.org>
wrote:

  
  


  


      

    

    

      
      

        


  

        

      Hervé Pagès
    

    

      
      #
    

  


  

      
Hi Burak,

Based on your description of what the package does, this is really a 
software package in nature, so is in the wrong repository. We need to 
fix this.

In 4 steps:

1. Please remove the ExperimentData term from the biocViews field as 
well as any term that belongs to the ExperimentData ontology, that is: 
DiseaseModel, OrganismData, Homo_sapiens_Data, Mus_musculus_Data, 
Rattus_norvegicus_Data, TechnologyData, MicroarrayData, SequencingData, 
and RNASeqData. Yep, most of your terms will go away.

2. Add the Software term, in first position. You're welcome to add new 
terms if you want, from the following vocabulary: 
https://github.com/Bioconductor/biocViews/blob/master/inst/dot/biocViewsVocab.dot. 
Please make sure to pick terms that belong to the Software ontology 
(i.e. are offsprings of the Software term).

3. Bump the package version (only the z part in x.y.z), and push your 
changes to git.bioconductor.org.

4. Let us know when you're done. We'll make the required adjustments on 
our side.

Thanks,

H.

        
On 28/09/2022 09:37, Burak Ogan Mancarci wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Hervé Pagès
    

    

      
      #
    

  


  

      
On 28/09/2022 16:03, Herv? Pag?s wrote:

        

            

      
      
      
    

        
After taking a second look I realize that **all** your current terms are 
from the ExperimentData ontology, instead of being a mix of Software and 
ExperimentData terms as I thought initially. So they should all be removed.

            

      
      
      
    

        
You should definitely add new terms from the Software ontology!

Thanks,

H.

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Burak Ogan Mancarci
    

    

      
      #
    

  


  

      
I have switched out the terms. Thanks a lot and sorry about the mess.

Cheers,
Ogan

On Wed, Sep 28, 2022 at 4:36 PM Herv? Pag?s <hpages.on.github at gmail.com>
wrote: