[BioC] code improvements

[Kuhn, Alexandre (NIH/NIA/IRP) [V]]

Mike,

if you have improvements to the code of annotationTools, you can contact me and I will be happy to try and incorporate them. My current email address is the package DESCRIPTION file that you can display using packageDescription("annotationTools").

You implied in your post on BioStar that column indices into annotation files are hard-coded in annotationTools functions. When designing them I actually tried to allow for flexibility in file format and provided the corresponding options, like e.g. for getGENEID

idCol: column in annotation table containing the gene IDs. 

Or for getHOMOLOG

clusterCol: column in homology/orthology table containing homology/orthology cluster IDs. 
speciesCol: column in homology/orthology table containing species IDs. 
idCol: column in homology/orthology table containing gene IDs.

I set the default values to match the format of Affymetrix files (for annotation functions) or HomoloGene file (for getHOMOLOG for instance). However these file formats might have changed since then. If this is the case, I will change the default values.

Alex


________________________________________
From: Vincent Carey [stvjc at channing.harvard.edu]
Sent: Monday, May 17, 2010 11:16 AM
To: Mike Dewar
Cc: bioconductor at stat.math.ethz.ch
Subject: Re: [BioC] code improvements

On Mon, May 17, 2010 at 10:49 AM, Mike Dewar
<mike.dewar at columbia.edu>wrote:

> Hello,
>
> I asked a question last week about how to submit patches.
>
> After hacking on the code ('annotationTools') for a while I realised (
> http://biostar.stackexchange.com/questions/1054/homology-bioconductor)
> that it was either the wrong package for the job or, what I think is
more
> likely, a package that had been largely superseded by another package
> ('biomaRt').
>
> So I have a new question, or maybe it's a suggestion: is there a way to
> mark packages for removal, or to add something to their entry in
> bioconductor that says "this package has been largely replaced by this
other
> package"?
>

We do remove packages.  When package functionalities decay, R supports a
Deprecated/Defunct annotation/user redirection process, and this is used
in
Bioconductor when appropriate.


>
> I should say that I feel rather mean asking this question, and it
shouldn't
> reflect badly on the authors of the package in question. It's just had
their
> been a pointer on the annotationTools package that said "these days, you
> should maybe check out 'biomaRt'" I would have saved myself quite a lot
of
> time. And I guess, in such a rapidly evolving area, and with such a
large
> code base, this sort of thing will be required sooner or later?
>
>
Your judgment on the duplication of annotationTools functionality by
biomaRt
seems to me overbroad.  The annotationTools vignette shows various
functionalities that are not supported by biomaRt.   I see no reason that
the two should not coexist.

Cheers,
>
> Mike Dewar
>
> On 13 May 2010, at 12:26, Vincent Carey wrote:
>
> The short answer is 'no', nothing official.  The long answer is that
> dealing with such improvements is up to the maintainer of the specific
> package affected.  If the maintainer is identified in the package
> DESCRIPTION, contact him/her directly.  If you don't see a maintainer
> identified (bioconductor devel team, for example, is listed as the
package
> maintainer) you could contact any of the developers directly.  For
example,
> you could write to me.
>
> On Wed, May 12, 2010 at 12:39 PM, Mike Dewar
<mike.dewar at columbia.edu>wrote:
>
>> Hello,
>>
>> If I make some improvements to a piece of code included in
Bioconductor,
>> is there an 'official' way of submitting the improvements for future
>> inclusion?
>>
>> Cheers,
>>
>> Mike Dewar
>>
>>
>> - - -
>> Dr Michael Dewar
>> Postdoctoral Research Scientist
>> Applied Mathematics
>> Columbia University
>> http://www.columbia.edu/~md2954/ <http://www.columbia.edu/%7Emd2954/>
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>
>
>
> - - -
> Dr Michael Dewar
> **Postdoctoral Research Scientist
> Applied Mathematics
> Columbia University
> http://www.columbia.edu/~md2954/ <http://www.columbia.edu/%7Emd2954/>
>
>
>
>
>
>
>

        [[alternative HTML version deleted]]

_______________________________________________
Bioconductor mailing list
Bioconductor at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives:
http://news.gmane.org/gmane.science.biology.informatics.conductor