Patchwork gnu: Add r-genomationdata.

login
register
mail settings
Submitter Ricardo Wurmus
Date May 23, 2016, 2:57 p.m.
Message ID <1464015436-27911-1-git-send-email-ricardo.wurmus@mdc-berlin.de>
Download mbox | patch
Permalink /patch/12462/
State New
Headers show

Comments

Ricardo Wurmus - May 23, 2016, 2:57 p.m.
* gnu/packages/bioinformatics.scm (r-genomationdata): New variable.
---
 gnu/packages/bioinformatics.scm | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)
Efraim Flashner - May 23, 2016, 6:09 p.m.
On Mon, May 23, 2016 at 04:57:16PM +0200, Ricardo Wurmus wrote:
> * gnu/packages/bioinformatics.scm (r-genomationdata): New variable.
> ---
>  gnu/packages/bioinformatics.scm | 26 ++++++++++++++++++++++++++
>  1 file changed, 26 insertions(+)
> 
> diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
> index fbaae12..51c371c 100644
> --- a/gnu/packages/bioinformatics.scm
> +++ b/gnu/packages/bioinformatics.scm
> @@ -5002,6 +5002,32 @@ genomic feature data as long as it has minimal information on the locations of
>  genomic intervals.  In addition, it can use BAM or BigWig files as input.")
>      (license license:artistic2.0)))
>  
> +(define-public r-genomationdata
> +  (package
> +    (name "r-genomationdata")
> +    (version "1.4.0")
> +    (source (origin
> +              (method url-fetch)
> +              ;; We cannot use bioconductor-uri here because this tarball is
> +              ;; located under "data/annotation/" instead of "bioc/".
> +              (uri (string-append "https://bioconductor.org/packages/"
> +                                  "release/data/experiment/src/contrib/"
> +                                  "genomationData_" version ".tar.gz"))
> +              (sha256
> +               (base32
> +                "1xzq2j722d8lcn5bc3aq3yb34xwis2d0bpsf6jsq4xw1bg0bsy79"))))
> +    (build-system r-build-system)
> +    (arguments `(#:substitutable? #f))
                    ^^^^
                why not?

> +    (native-inputs
> +     `(("r-knitr" ,r-knitr)))
> +    (home-page "http://bioinformatics.mdc-berlin.de/genomation/")
> +    (synopsis "Experimental data for use with the genomation package")
> +    (description
> +     "This package contains experimental genetic data for use with the
> +genomation package.  Included are Chip Seq, Methylation and Cage data,
> +downloaded from Encode.")
> +    (license license:gpl3+)))
> +
>  (define-public r-org-hs-eg-db
>    (package
>      (name "r-org-hs-eg-db")
> -- 
> 2.7.3
> 

Looks good
Ricardo Wurmus - May 24, 2016, 6:36 a.m.
Efraim Flashner <efraim@flashner.co.il> writes:

> On Mon, May 23, 2016 at 04:57:16PM +0200, Ricardo Wurmus wrote:
>> * gnu/packages/bioinformatics.scm (r-genomationdata): New variable.
>> ---

[...]

>> +    (arguments `(#:substitutable? #f))
>                     ^^^^
>                 why not?

Because it’s an R data package, i.e. it’s a large download taking up
space on hydra; it’s the same for all architechtures and there’s no real
advantage in providing a binary substitute.

We’ve done the same for the genome data packages.  It was more clear cut
then, because the download was about ten times bigger.

Still, I didn’t want to waste space on hydra with a substitute for
something that’s mostly a simple download.

~~ Ricardo
Efraim Flashner - May 24, 2016, 7:09 a.m.
On Tue, May 24, 2016 at 08:36:22AM +0200, Ricardo Wurmus wrote:
> 
> Efraim Flashner <efraim@flashner.co.il> writes:
> 
> > On Mon, May 23, 2016 at 04:57:16PM +0200, Ricardo Wurmus wrote:
> >> * gnu/packages/bioinformatics.scm (r-genomationdata): New variable.
> >> ---
> 
> [...]
> 
> >> +    (arguments `(#:substitutable? #f))
> >                     ^^^^
> >                 why not?
> 
> Because it’s an R data package, i.e. it’s a large download taking up
> space on hydra; it’s the same for all architechtures and there’s no real
> advantage in providing a binary substitute.
> 
> We’ve done the same for the genome data packages.  It was more clear cut
> then, because the download was about ten times bigger.
> 
> Still, I didn’t want to waste space on hydra with a substitute for
> something that’s mostly a simple download.
> 
> ~~ Ricardo

I was a little short with words, what I meant to write was "we add a
comment when we have #:tests? #f, should we add a comment here too?"
Ludovic Courtès - May 24, 2016, 9:52 a.m.
Ricardo Wurmus <ricardo.wurmus@mdc-berlin.de> skribis:

> Efraim Flashner <efraim@flashner.co.il> writes:
>
>> On Mon, May 23, 2016 at 04:57:16PM +0200, Ricardo Wurmus wrote:
>>> * gnu/packages/bioinformatics.scm (r-genomationdata): New variable.
>>> ---
>
> [...]
>
>>> +    (arguments `(#:substitutable? #f))
>>                     ^^^^
>>                 why not?
>
> Because it’s an R data package, i.e. it’s a large download taking up
> space on hydra; it’s the same for all architechtures and there’s no real
> advantage in providing a binary substitute.

Makes sense.  Maybe just mention it in a comment above.

Thanks,
Ludo’.
Ricardo Wurmus - May 31, 2016, 2:30 p.m.
Ludovic Courtès <ludo@gnu.org> writes:

> Ricardo Wurmus <ricardo.wurmus@mdc-berlin.de> skribis:
>
>> Efraim Flashner <efraim@flashner.co.il> writes:
>>
>>> On Mon, May 23, 2016 at 04:57:16PM +0200, Ricardo Wurmus wrote:
>>>> * gnu/packages/bioinformatics.scm (r-genomationdata): New variable.
>>>> ---
>>
>> [...]
>>
>>>> +    (arguments `(#:substitutable? #f))
>>>                     ^^^^
>>>                 why not?
>>
>> Because it’s an R data package, i.e. it’s a large download taking up
>> space on hydra; it’s the same for all architechtures and there’s no real
>> advantage in providing a binary substitute.
>
> Makes sense.  Maybe just mention it in a comment above.

I added a comment and pushed.  Thanks!

~~ Ricardo

Patch

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index fbaae12..51c371c 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -5002,6 +5002,32 @@  genomic feature data as long as it has minimal information on the locations of
 genomic intervals.  In addition, it can use BAM or BigWig files as input.")
     (license license:artistic2.0)))
 
+(define-public r-genomationdata
+  (package
+    (name "r-genomationdata")
+    (version "1.4.0")
+    (source (origin
+              (method url-fetch)
+              ;; We cannot use bioconductor-uri here because this tarball is
+              ;; located under "data/annotation/" instead of "bioc/".
+              (uri (string-append "https://bioconductor.org/packages/"
+                                  "release/data/experiment/src/contrib/"
+                                  "genomationData_" version ".tar.gz"))
+              (sha256
+               (base32
+                "1xzq2j722d8lcn5bc3aq3yb34xwis2d0bpsf6jsq4xw1bg0bsy79"))))
+    (build-system r-build-system)
+    (arguments `(#:substitutable? #f))
+    (native-inputs
+     `(("r-knitr" ,r-knitr)))
+    (home-page "http://bioinformatics.mdc-berlin.de/genomation/")
+    (synopsis "Experimental data for use with the genomation package")
+    (description
+     "This package contains experimental genetic data for use with the
+genomation package.  Included are Chip Seq, Methylation and Cage data,
+downloaded from Encode.")
+    (license license:gpl3+)))
+
 (define-public r-org-hs-eg-db
   (package
     (name "r-org-hs-eg-db")