[PATCH 00/18] Add R packages

[PATCH 00/18] Add R packages

[Ricardo Wurmus]

Hi Guix,

what follows are a couple of R packages that I've been using at work for a few
weeks.  To avoid delays I often add requested packages to our local repository
first; when my staging file gets too big I take some time to upstream them.

All these patches have been checked with "guix lint".

These are all the R packages I currently have in my staging file.  A few more
bioinfo things might follow later.

~~ Ricardo

Ricardo Wurmus (18):
  gnu: Add r-rematch.
  gnu: Add r-cellranger.
  gnu: Add r-googlesheets.
  gnu: Add r-spams.
  gnu: Add r-r4rna.
  gnu: Add r-dynamictreecut.
  gnu: Add r-preprocesscore.
  gnu: Add r-fastcluster.
  gnu: Add r-wgcna.
  gnu: Add r-sfsmisc.
  gnu: Add r-kernlab.
  gnu: Add r-gtools.
  gnu: Add r-chipkernels.
  gnu: Add r-seqgl.
  gnu: Add r-gdata.
  gnu: Add r-gplots.
  gnu: Add r-rocr.
  gnu: Add r-gkmsvm.

 gnu/packages/bioinformatics.scm   | 158 ++++++++++++++++++
 gnu/packages/machine-learning.scm |  22 +++
 gnu/packages/statistics.scm       | 340 ++++++++++++++++++++++++++++++++++++++
 3 files changed, 520 insertions(+)

-- 
2.10.2

[PATCH 01/18] gnu: Add r-rematch.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-rematch): New variable.
---
 gnu/packages/statistics.scm | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index 700a895..612ea53 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3272,6 +3272,26 @@ noncentral hypergeometric distribution (also called extended hypergeometric
 distribution).")
    (license license:gpl3+)))
 
+(define-public r-rematch
+  (package
+    (name "r-rematch")
+    (version "1.0.1")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "rematch" version))
+       (sha256
+        (base32
+         "0y3mshvpvz9csmq8hk8jbabx4nxlv5sckvfzvm6920ndg34xw2d4"))))
+    (build-system r-build-system)
+    (home-page "https://github.com/MangoTheCat/rematch")
+    (synopsis "Match regular expressions with a nicer API")
+    (description
+     "This package provides a small wrapper on @code{regexpr} to extract the
+matches and captured groups from the match of a regular expression to a
+character vector.")
+    (license license:expat)))
+
 (define-public r-rpart
   (package
     (name "r-rpart")
-- 
2.10.2

[PATCH 02/18] gnu: Add r-cellranger.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-cellranger): New variable.
---
 gnu/packages/statistics.scm | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index 612ea53..cec304d 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3292,6 +3292,28 @@ matches and captured groups from the match of a regular expression to a
 character vector.")
     (license license:expat)))
 
+(define-public r-cellranger
+  (package
+    (name "r-cellranger")
+    (version "1.1.0")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "cellranger" version))
+       (sha256
+        (base32
+         "16fgi3annn34c3cxi0pxf62mmmmxi21hp0zzlv7bkfsjqy4g4f2x"))))
+    (build-system r-build-system)
+    (propagated-inputs
+     `(("r-rematch" ,r-rematch)
+       ("r-tibble" ,r-tibble)))
+    (home-page "https://github.com/rsheets/cellranger")
+    (synopsis "Translate spreadsheet cell ranges to rows and columns")
+    (description
+     "This package provides helper functions to work with spreadsheets and the
+@code{A1:D10} style of cell range specification.")
+    (license license:expat)))
+
 (define-public r-rpart
   (package
     (name "r-rpart")
-- 
2.10.2

[PATCH 03/18] gnu: Add r-googlesheets.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-googlesheets): New variable.
---
 gnu/packages/statistics.scm | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index cec304d..0ba3316 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3314,6 +3314,34 @@ character vector.")
 @code{A1:D10} style of cell range specification.")
     (license license:expat)))
 
+(define-public r-googlesheets
+  (package
+    (name "r-googlesheets")
+    (version "0.2.1")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "googlesheets" version))
+       (sha256
+        (base32
+         "0ps13h1cv7fj5dh8s4nvwi64wnnyqdsadcaa4iizq1c5s615cwk3"))))
+    (build-system r-build-system)
+    (propagated-inputs
+     `(("r-cellranger" ,r-cellranger)
+       ("r-dplyr" ,r-dplyr)
+       ("r-httr" ,r-httr)
+       ("r-jsonlite" ,r-jsonlite)
+       ("r-purrr" ,r-purrr)
+       ("r-readr" ,r-readr)
+       ("r-stringr" ,r-stringr)
+       ("r-tidyr" ,r-tidyr)
+       ("r-xml2" ,r-xml2)))
+    (home-page "https://github.com/jennybc/googlesheets")
+    (synopsis "Manage Google spreadsheets from R")
+    (description "This package provides tools to interact with Google Sheets
+from within R.")
+    (license license:expat)))
+
 (define-public r-rpart
   (package
     (name "r-rpart")
-- 
2.10.2

[PATCH 04/18] gnu: Add r-spams.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-spams): New variable.
---
 gnu/packages/statistics.scm | 52 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 52 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index 0ba3316..dcf6dbe 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3342,6 +3342,58 @@ character vector.")
 from within R.")
     (license license:expat)))
 
+(define-public r-spams
+  (package
+    (name "r-spams")
+    (version "2.5-svn2014-07-04")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (string-append "https://gforge.inria.fr/frs/download.php/33815/"
+                           "spams-R-v" version ".tar.gz"))
+       (sha256
+        (base32
+         "1k459jg9a334slkw31w63l4d39xszjzsng7dv5j1mp78zifz7hvx"))))
+    (build-system r-build-system)
+    (arguments
+     `(#:phases
+       (modify-phases %standard-phases
+         (add-after 'unpack 'chdir
+           (lambda _ (chdir "spams") #t))
+         ;; Since R 3.3.0 including R headers inside of an extern "C" block
+         ;; causes C headers to be included, which results in a lot of
+         ;; duplicate definitions.  This can be avoided by defining
+         ;; NO_C_HEADERS before including the R headers.
+         (add-after 'chdir 'patch-use-of-R-headers
+           (lambda _
+             (substitute* "src/spams.cpp"
+               (("#include <R.h>" line)
+                (string-append "#define NO_C_HEADERS\n" line)))
+             #t))
+         ;; This looks like a syntax error.
+         (add-after 'chdir 'patch-isnan
+           (lambda _
+             (substitute* '"src/spams/linalg/linalg.h"
+               (("if isnan\\(lambda\\) \\{")
+                "if (isnan(lambda)) {"))
+             #t)))))
+    (home-page "http://spams-devel.gforge.inria.fr")
+    (synopsis "Toolbox for solving sparse estimation problems")
+    (description "SPAMS (SPArse Modeling Software) is an optimization toolbox
+for solving various sparse estimation problems.  It includes tools for the
+following problems:
+
+@enumerate
+@item Dictionary learning and matrix factorization (NMF, sparse @dfn{principle
+ component analysis} (PCA), ...)
+@item Solving sparse decomposition problems with LARS, coordinate descent,
+ OMP, SOMP, proximal methods
+@item Solving structured sparse decomposition problems (l1/l2, l1/linf, sparse
+ group lasso, tree-structured regularization, structured sparsity with
+ overlapping groups,...).
+@end enumerate\n")
+    (license license:gpl3+)))
+
 (define-public r-rpart
   (package
     (name "r-rpart")
-- 
2.10.2

Re: [PATCH 04/18] gnu: Add r-spams.

[Leo Famulari]

On Thu, Nov 24, 2016 at 05:51:25PM +0100, Ricardo Wurmus wrote:
> * gnu/packages/statistics.scm (r-spams): New variable.

> +         ;; This looks like a syntax error.
> +         (add-after 'chdir 'patch-isnan
> +           (lambda _
> +             (substitute* '"src/spams/linalg/linalg.h"
> +               (("if isnan\\(lambda\\) \\{")
> +                "if (isnan(lambda)) {"))
> +             #t)))))

Has this been reported upstream? :)

LGTM regardless.

Re: [PATCH 04/18] gnu: Add r-spams.

[Ricardo Wurmus]

Leo Famulari <leo@famulari.name> writes:

> On Thu, Nov 24, 2016 at 05:51:25PM +0100, Ricardo Wurmus wrote:
>> * gnu/packages/statistics.scm (r-spams): New variable.
>
>> +         ;; This looks like a syntax error.
>> +         (add-after 'chdir 'patch-isnan
>> +           (lambda _
>> +             (substitute* '"src/spams/linalg/linalg.h"
>> +               (("if isnan\\(lambda\\) \\{")
>> +                "if (isnan(lambda)) {"))
>> +             #t)))))
>
> Has this been reported upstream? :)

I’ve just sent an email to the developer.

-- 
Ricardo

GPG: BCA6 89B6 3655 3801 C3C6  2150 197A 5888 235F ACAC
http://elephly.net

[PATCH 05/18] gnu: Add r-r4rna.

[Ricardo Wurmus]

* gnu/packages/bioinformatics.scm (r-r4rna): New variable.
---
 gnu/packages/bioinformatics.scm | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index 773b590..254ee13 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -7246,6 +7246,29 @@ two-dimensional genome scans.")
 libraries for systems that do not have these available via other means.")
     (license license:artistic2.0)))
 
+(define-public r-r4rna
+  (package
+    (name "r-r4rna")
+    (version "0.1.4")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (string-append "http://www.e-rna.org/r-chie/files/R4RNA_"
+                           version ".tar.gz"))
+       (sha256
+        (base32
+         "1p0i78wh76jfgmn9jphbwwaz6yy6pipzfg08xs54cxavxg2j81p5"))))
+    (build-system r-build-system)
+    (propagated-inputs
+     `(("r-optparse" ,r-optparse)
+       ("r-rcolorbrewer" ,r-rcolorbrewer)))
+    (home-page "http://www.e-rna.org/r-chie/index.cgi")
+    (synopsis "Analysis framework for RNA secondary structure")
+    (description
+     "The R4RNA package aims to be a general framework for the analysis of RNA
+secondary structure and comparative analysis in R.")
+    (license license:gpl3+)))
+
 (define-public r-rhtslib
   (package
     (name "r-rhtslib")
-- 
2.10.2

[PATCH 06/18] gnu: Add r-dynamictreecut.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-dynamictreecut): New variable.
---
 gnu/packages/statistics.scm | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index dcf6dbe..c486208 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3497,6 +3497,27 @@ conversion of R objects to LaTeX code, and recoding variables.")
 framework, with additional code inspection and report generation tools.")
     (license license:gpl2+)))
 
+(define-public r-dynamictreecut
+  (package
+    (name "r-dynamictreecut")
+    (version "1.63-1")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "dynamicTreeCut" version))
+       (sha256
+        (base32
+         "1fadbql7g5r2vvlkr89nlrjxwp4yx4xrdqmv077qvmnx9vv0f4w3"))))
+    (properties `((upstream-name . "dynamicTreeCut")))
+    (build-system r-build-system)
+    (home-page
+     "http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork/BranchCutting/")
+    (synopsis "Detect clusters in hierarchical clustering dendrograms")
+    (description
+     "This package contains methods for the detection of clusters in
+hierarchical clustering dendrograms.")
+    (license license:gpl2+)))
+
 (define-public r-kernsmooth
   (package
     (name "r-kernsmooth")
-- 
2.10.2

[PATCH 07/18] gnu: Add r-preprocesscore.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-preprocesscore): New variable.
---
 gnu/packages/statistics.scm | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index c486208..d2ca35e 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3518,6 +3518,27 @@ framework, with additional code inspection and report generation tools.")
 hierarchical clustering dendrograms.")
     (license license:gpl2+)))
 
+(define-public r-preprocesscore
+  (package
+    (name "r-preprocesscore")
+    (version "1.36.0")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (bioconductor-uri "preprocessCore" version))
+       (sha256
+        (base32
+         "1n8y12q7145f385gm2k3c6y3vwvin7jlb47la4mnl7mar6pq9kmp"))))
+    (properties
+     `((upstream-name . "preprocessCore")))
+    (build-system r-build-system)
+    (home-page "https://github.com/bmbolstad/preprocessCore")
+    (synopsis "Collection of pre-processing functions")
+    (description
+     "This package provides a library of core pre-processing and normalization
+routines.")
+    (license license:lgpl2.0+)))
+
 (define-public r-kernsmooth
   (package
     (name "r-kernsmooth")
-- 
2.10.2

[PATCH 08/18] gnu: Add r-fastcluster.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-fastcluster): New variable.
---
 gnu/packages/statistics.scm | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index d2ca35e..3c2bd0c 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3539,6 +3539,31 @@ hierarchical clustering dendrograms.")
 routines.")
     (license license:lgpl2.0+)))
 
+(define-public r-fastcluster
+  (package
+    (name "r-fastcluster")
+    (version "1.1.20")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "fastcluster" version))
+       (sha256
+        (base32
+         "0rlbxhh894znf10x0xgkv9dzpibgq9jw5aqpgviccdnxc2c5hwid"))))
+    (build-system r-build-system)
+    (home-page "http://danifold.net/fastcluster.html")
+    (synopsis "Fast hierarchical clustering routines")
+    (description
+     "This package implements fast hierarchical, agglomerative clustering
+routines.  Part of the functionality is designed as drop-in replacement for
+existing routines: @code{linkage()} in the SciPy package
+@code{scipy.cluster.hierarchy}, @code{hclust()} in R's @code{stats} package,
+and the @code{flashClust} package.  It provides the same functionality with
+the benefit of a much faster implementation.  Moreover, there are
+memory-saving routines for clustering of vector data, which go beyond what the
+existing packages provide.")
+    (license license:bsd-2)))
+
 (define-public r-kernsmooth
   (package
     (name "r-kernsmooth")
-- 
2.10.2

[PATCH 09/18] gnu: Add r-wgcna.

[Ricardo Wurmus]

* gnu/packages/bioinformatics.scm (r-wgcna): New variable.
---
 gnu/packages/bioinformatics.scm | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index 254ee13..7bcac59 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -7449,6 +7449,42 @@ characterization and visualization of a wide range of mutational patterns
 in SNV base substitution data.")
     (license license:expat)))
 
+(define-public r-wgcna
+  (package
+    (name "r-wgcna")
+    (version "1.51")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "WGCNA" version))
+       (sha256
+        (base32
+         "0hzvnhw76vwg8bl8x368f0c5szpwb8323bmrb3bir93i5bmfjsxx"))))
+    (properties `((upstream-name . "WGCNA")))
+    (build-system r-build-system)
+    (propagated-inputs
+     `(("r-annotationdbi" ,r-annotationdbi)
+       ("r-doparallel" ,r-doparallel)
+       ("r-dynamictreecut" ,r-dynamictreecut)
+       ("r-fastcluster" ,r-fastcluster)
+       ("r-foreach" ,r-foreach)
+       ("r-go-db" ,r-go-db)
+       ("r-hmisc" ,r-hmisc)
+       ("r-impute" ,r-impute)
+       ("r-matrixstats" ,r-matrixstats)
+       ("r-preprocesscore" ,r-preprocesscore)))
+    (home-page
+     "http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork/Rpackages/WGCNA/")
+    (synopsis "Weighted correlation network analysis")
+    (description
+     "This package provides functions necessary to perform Weighted
+Correlation Network Analysis on high-dimensional data.  It includes functions
+for rudimentary data cleaning, construction of correlation networks, module
+identification, summarization, and relating of variables and modules to sample
+traits.  It also includes a number of utility functions for data manipulation
+and visualization.")
+    (license license:gpl2+)))
+
 (define-public emboss
   (package
     (name "emboss")
-- 
2.10.2

Re: [PATCH 09/18] gnu: Add r-wgcna.

[Ben Woodcroft]

On 25/11/16 02:51, Ricardo Wurmus wrote:
> [..]
> +    (home-page
> +     "http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork/Rpackages/WGCNA/")
This line is a bit long, I don't remember, should it be split over two 
lines?

> +    (synopsis "Weighted correlation network analysis")
> +    (description
> +     "This package provides functions necessary to perform Weighted
> +Correlation Network Analysis on high-dimensional data.  It includes functions
> +for rudimentary data cleaning, construction of correlation networks, module
> +identification, summarization, and relating of variables and modules to sample
I found the 'and' a bit confusing. Maybe:

It includes functions for rudimentary data cleaning, construction and 
summarization of correlation networks, module identification and 
functions for relating both variables and modules to sample traits.

> +traits.  It also includes a number of utility functions for data manipulation
> +and visualization.")
> +    (license license:gpl2+)))
> +
>   (define-public emboss
>     (package
>       (name "emboss")
LGTM, I have one just like it locally.
ben

Re: [PATCH 09/18] gnu: Add r-wgcna.

[Ricardo Wurmus]

Ben Woodcroft <b.woodcroft@uq.edu.au> writes:

> On 25/11/16 02:51, Ricardo Wurmus wrote:
>> [..]
>> +    (home-page
>> +     "http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork/Rpackages/WGCNA/")
> This line is a bit long, I don't remember, should it be split over two 
> lines?

True, but I left it as it is.  It’s just a tiny bit too long and we
don’t usually split home-page strings, so I opted to keep it like this.

>> +    (synopsis "Weighted correlation network analysis")
>> +    (description
>> +     "This package provides functions necessary to perform Weighted
>> +Correlation Network Analysis on high-dimensional data.  It includes functions
>> +for rudimentary data cleaning, construction of correlation networks, module
>> +identification, summarization, and relating of variables and modules to sample
> I found the 'and' a bit confusing. Maybe:
>
> It includes functions for rudimentary data cleaning, construction and 
> summarization of correlation networks, module identification and 
> functions for relating both variables and modules to sample traits.

Thanks for the improved sentence.  I’ve updated the description.

>> +traits.  It also includes a number of utility functions for data manipulation
>> +and visualization.")
>> +    (license license:gpl2+)))
>> +
>>   (define-public emboss
>>     (package
>>       (name "emboss")
> LGTM, I have one just like it locally.
> ben

Cool, thanks for the review!

-- 
Ricardo

GPG: BCA6 89B6 3655 3801 C3C6  2150 197A 5888 235F ACAC
http://elephly.net

[PATCH 10/18] gnu: Add r-sfsmisc.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-sfsmisc): New variable.
---
 gnu/packages/statistics.scm | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index 3c2bd0c..0c54258 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3564,6 +3564,25 @@ memory-saving routines for clustering of vector data, which go beyond what the
 existing packages provide.")
     (license license:bsd-2)))
 
+(define-public r-sfsmisc
+  (package
+    (name "r-sfsmisc")
+    (version "1.1-0")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "sfsmisc" version))
+       (sha256
+        (base32
+         "0580piv4n1nispl3pa8nfjjfnb8iwaqky2dzdy0aqnxrxgrhqhvz"))))
+    (build-system r-build-system)
+    (home-page "http://cran.r-project.org/web/packages/sfsmisc")
+    (synopsis "Utilities from \"Seminar fuer Statistik\" ETH Zurich")
+    (description
+     "This package provides useful utilities from Seminar fuer Statistik ETH
+Zurich, including many that are related to graphics.")
+    (license license:gpl2+)))
+
 (define-public r-kernsmooth
   (package
     (name "r-kernsmooth")
-- 
2.10.2

[PATCH 11/18] gnu: Add r-kernlab.

[Ricardo Wurmus]

* gnu/packages/machine-learning.scm (r-kernlab): New variable.
---
 gnu/packages/machine-learning.scm | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/gnu/packages/machine-learning.scm b/gnu/packages/machine-learning.scm
index c239c4f..8f1f8ee 100644
--- a/gnu/packages/machine-learning.scm
+++ b/gnu/packages/machine-learning.scm
@@ -479,6 +479,28 @@ geometric models.")
 single hidden layer, and for multinomial log-linear models.")
     (license (list license:gpl2+ license:gpl3+))))
 
+(define-public r-kernlab
+  (package
+    (name "r-kernlab")
+    (version "0.9-25")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "kernlab" version))
+       (sha256
+        (base32
+         "0qnaq9x3j2xc6jrmmd98wc6hkzch487s4p3a9lnc00xvahkhgpmr"))))
+    (build-system r-build-system)
+    (home-page "http://cran.r-project.org/web/packages/kernlab")
+    (synopsis "Kernel-based machine learning tools")
+    (description
+     "This package provides kernel-based machine learning methods for
+classification, regression, clustering, novelty detection, quantile regression
+and dimensionality reduction.  Among other methods @code{kernlab} includes
+Support Vector Machines, Spectral Clustering, Kernel PCA, Gaussian Processes
+and a QP solver.")
+    (license license:gpl2)))
+
 (define-public dlib
   (package
     (name "dlib")
-- 
2.10.2

[PATCH 12/18] gnu: Add r-gtools.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-gtools): New variable.
---
 gnu/packages/statistics.scm | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index 0c54258..8027aed 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3583,6 +3583,28 @@ existing packages provide.")
 Zurich, including many that are related to graphics.")
     (license license:gpl2+)))
 
+(define-public r-gtools
+  (package
+    (name "r-gtools")
+    (version "3.5.0")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "gtools" version))
+       (sha256
+        (base32
+         "1xknwk9xlsj027pg0nwiizigcrsc84hdrig0jn0cgcyxj8dabdl6"))))
+    (build-system r-build-system)
+    (home-page "http://cran.r-project.org/web/packages/gtools")
+    (synopsis "Various R programming tools")
+    (description
+     "This package contains a collection of various functions to assist in R
+programming, such as tools to assist in developing, updating, and maintaining
+R and R packages, calculating the logit and inverse logit transformations,
+tests for whether a value is missing, empty or contains only @code{NA} and
+@code{NULL} values, and many more.")
+    (license license:gpl2)))
+
 (define-public r-kernsmooth
   (package
     (name "r-kernsmooth")
-- 
2.10.2

[PATCH 13/18] gnu: Add r-chipkernels.

[Ricardo Wurmus]

* gnu/packages/bioinformatics.scm (r-chipkernels): New variable.
---
 gnu/packages/bioinformatics.scm | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index 7bcac59..98dad40 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -7485,6 +7485,42 @@ traits.  It also includes a number of utility functions for data manipulation
 and visualization.")
     (license license:gpl2+)))
 
+(define-public r-chipkernels
+  (let ((commit "c9cfcacb626b1221094fb3490ea7bac0fd625372")
+        (revision "1"))
+    (package
+      (name "r-chipkernels")
+      (version (string-append "1.1-" revision "." (string-take commit 9)))
+      (source
+       (origin
+         (method git-fetch)
+         (uri (git-reference
+               (url "https://github.com/ManuSetty/ChIPKernels.git")
+               (commit commit)))
+         (file-name (string-append name "-" version))
+         (sha256
+          (base32
+           "14bj5qhjm1hsm9ay561nfbqi9wxsa7y487df2idsaaf6z10nw4v0"))))
+      (build-system r-build-system)
+      (propagated-inputs
+       `(("r-iranges" ,r-iranges)
+         ("r-xvector" ,r-xvector)
+         ("r-biostrings" ,r-biostrings)
+         ("r-bsgenome" ,r-bsgenome)
+         ("r-gtools" ,r-gtools)
+         ("r-genomicranges" ,r-genomicranges)
+         ("r-sfsmisc" ,r-sfsmisc)
+         ("r-kernlab" ,r-kernlab)
+         ("r-s4vectors" ,r-s4vectors)
+         ("r-biocgenerics" ,r-biocgenerics)))
+      (home-page "https://github.com/ManuSetty/ChIPKernels")
+      (synopsis "Build string kernels for DNA Sequence analysis")
+      (description "ChIPKernels is an R package for building different string
+kernels used for DNA Sequence analysis.  A dictionary of the desired kernel
+must be built and this dictionary can be used for determining kernels for DNA
+Sequences.")
+      (license license:gpl2+))))
+
 (define-public emboss
   (package
     (name "emboss")
-- 
2.10.2

[PATCH 14/18] gnu: Add r-seqgl.

[Ricardo Wurmus]

* gnu/packages/bioinformatics.scm (r-seqgl): New variable.
---
 gnu/packages/bioinformatics.scm | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index 98dad40..89a9db1 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -7521,6 +7521,36 @@ must be built and this dictionary can be used for determining kernels for DNA
 Sequences.")
       (license license:gpl2+))))
 
+(define-public r-seqgl
+  (package
+    (name "r-seqgl")
+    (version "1.1.4")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (string-append "https://github.com/ManuSetty/SeqGL/"
+                           "archive/" version ".tar.gz"))
+       (file-name (string-append name "-" version ".tar.gz"))
+       (sha256
+        (base32
+         "0pnk1p3sci5yipyc8xnb6jbmydpl80fld927xgnbcv104hy8h8yh"))))
+    (build-system r-build-system)
+    (propagated-inputs
+     `(("r-biostrings" ,r-biostrings)
+       ("r-chipkernels" ,r-chipkernels)
+       ("r-genomicranges" ,r-genomicranges)
+       ("r-spams" ,r-spams)
+       ("r-wgcna" ,r-wgcna)
+       ("r-fastcluster" ,r-fastcluster)))
+    (home-page "https://github.com/ManuSetty/SeqGL")
+    (synopsis "Group lasso for Dnase/ChIP-seq data")
+    (description "SeqGL is a group lasso based algorithm to extract
+transcription factor sequence signals from ChIP, DNase and ATAC-seq profiles.
+This package presents a method which uses group lasso to discriminate between
+bound and non bound genomic regions to accurately identify transcription
+factors bound at the specific regions.")
+    (license license:gpl2+)))
+
 (define-public emboss
   (package
     (name "emboss")
-- 
2.10.2

[PATCH 15/18] gnu: Add r-gdata.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-gdata): New variable.
---
 gnu/packages/statistics.scm | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index 8027aed..dc519b4 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3605,6 +3605,46 @@ tests for whether a value is missing, empty or contains only @code{NA} and
 @code{NULL} values, and many more.")
     (license license:gpl2)))
 
+(define-public r-gdata
+  (package
+    (name "r-gdata")
+    (version "2.17.0")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "gdata" version))
+       (sha256
+        (base32
+         "0kiy3jbcszlpmarg311spdsfi5pn89wgy742dxsbzxk8907fr5w0"))))
+    (build-system r-build-system)
+    (inputs
+     `(("perl" ,perl)))
+    (propagated-inputs
+     `(("r-gtools" ,r-gtools)))
+    (home-page "http://cran.r-project.org/web/packages/gdata")
+    (synopsis "Various R programming tools for data manipulation")
+    (description
+     "This package provides various R programming tools for data manipulation,
+including:
+
+@itemize
+@item medical unit conversions
+@item combining objects
+@item character vector operations
+@item factor manipulation
+@item obtaining information about R objects
+@item manipulating MS-Excel formatted files
+@item generating fixed-width format files
+@item extricating components of date and time objects
+@item operations on columns of data frames
+@item matrix operations
+@item operations on vectors and data frames
+@item value of last evaluated expression
+@item wrapper for @code{sample} that ensures consistent behavior for
+  both scalar and vector arguments
+@end itemize\n")
+    (license license:gpl2+)))
+
 (define-public r-kernsmooth
   (package
     (name "r-kernsmooth")
-- 
2.10.2

Re: [PATCH 15/18] gnu: Add r-gdata.

[Leo Famulari]

On Thu, Nov 24, 2016 at 05:51:36PM +0100, Ricardo Wurmus wrote:
> * gnu/packages/statistics.scm (r-gdata): New variable.

> +    (description
> +     "This package provides various R programming tools for data manipulation,
> +including:
> +
> +@itemize
> +@item medical unit conversions
> +@item combining objects
> +@item character vector operations
> +@item factor manipulation
> +@item obtaining information about R objects
> +@item manipulating MS-Excel formatted files
> +@item generating fixed-width format files
> +@item extricating components of date and time objects
> +@item operations on columns of data frames
> +@item matrix operations
> +@item operations on vectors and data frames
> +@item value of last evaluated expression
> +@item wrapper for @code{sample} that ensures consistent behavior for
> +  both scalar and vector arguments
> +@end itemize\n")

Same question about the texinfo markup as for r-gplots.

[PATCH 16/18] gnu: Add r-gplots.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-gplots): New variable.
---
 gnu/packages/statistics.scm | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index dc519b4..f014f16 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3645,6 +3645,46 @@ including:
 @end itemize\n")
     (license license:gpl2+)))
 
+(define-public r-gplots
+  (package
+    (name "r-gplots")
+    (version "3.0.1")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "gplots" version))
+       (sha256
+        (base32
+         "02nb8n3s7c1zxq2s7ycaq2ys72y7mzirxrwj954h6gdc4x1zhg9l"))))
+    (build-system r-build-system)
+    (propagated-inputs
+     `(("r-catools" ,r-catools)
+       ("r-gdata" ,r-gdata)
+       ("r-gtools" ,r-gtools)
+       ("r-kernsmooth" ,r-kernsmooth)))
+    (home-page "http://cran.r-project.org/web/packages/gplots")
+    (synopsis "Various R programming tools for plotting data")
+    (description
+     "This package provides various R programming tools for plotting data,
+including:
+
+@itemize
+@item calculating and plotting locally smoothed summary function
+@item enhanced versions of standard plots
+@item manipulating colors
+@item calculating and plotting two-dimensional data summaries
+@item enhanced regression diagnostic plots
+@item formula-enabled interface to @code{stats::lowess} function
+@item displaying textual data in plots
+@item baloon plots
+@item plotting \"Venn\" diagrams
+@item displaying Open-Office style plots
+@item plotting multiple data on same region, with separate axes
+@item plotting means and confidence intervals
+@item spacing points in an x-y plot so they don't overlap
+@end itemize\n")
+    (license license:gpl2+)))
+
 (define-public r-kernsmooth
   (package
     (name "r-kernsmooth")
-- 
2.10.2

Re: [PATCH 16/18] gnu: Add r-gplots.

[Leo Famulari]

On Thu, Nov 24, 2016 at 05:51:37PM +0100, Ricardo Wurmus wrote:
> * gnu/packages/statistics.scm (r-gplots): New variable.

> +    (description
> +     "This package provides various R programming tools for plotting data,
> +including:
> +
> +@itemize
> +@item calculating and plotting locally smoothed summary function
> +@item enhanced versions of standard plots
> +@item manipulating colors
> +@item calculating and plotting two-dimensional data summaries
> +@item enhanced regression diagnostic plots
> +@item formula-enabled interface to @code{stats::lowess} function
> +@item displaying textual data in plots
> +@item baloon plots
> +@item plotting \"Venn\" diagrams
> +@item displaying Open-Office style plots
> +@item plotting multiple data on same region, with separate axes
> +@item plotting means and confidence intervals
> +@item spacing points in an x-y plot so they don't overlap
> +@end itemize\n")

We had been using @enumerate instead of @itemize to work around
<http://bugs.gnu.org/21772>. Do you know if Andy's fix is in our Guile
yet?

Re: [PATCH 16/18] gnu: Add r-gplots.

[Ludovic Courtès]

Leo Famulari <leo@famulari.name> skribis:

> On Thu, Nov 24, 2016 at 05:51:37PM +0100, Ricardo Wurmus wrote:
>> * gnu/packages/statistics.scm (r-gplots): New variable.
>
>> +    (description
>> +     "This package provides various R programming tools for plotting data,
>> +including:
>> +
>> +@itemize
>> +@item calculating and plotting locally smoothed summary function
>> +@item enhanced versions of standard plots
>> +@item manipulating colors
>> +@item calculating and plotting two-dimensional data summaries
>> +@item enhanced regression diagnostic plots
>> +@item formula-enabled interface to @code{stats::lowess} function
>> +@item displaying textual data in plots
>> +@item baloon plots
>> +@item plotting \"Venn\" diagrams
>> +@item displaying Open-Office style plots
>> +@item plotting multiple data on same region, with separate axes
>> +@item plotting means and confidence intervals
>> +@item spacing points in an x-y plot so they don't overlap
>> +@end itemize\n")
>
> We had been using @enumerate instead of @itemize to work around
> <http://bugs.gnu.org/21772>. Do you know if Andy's fix is in our Guile
> yet?

Yes, it’s in 2.0.12.

Ludo’.

Re: [PATCH 16/18] gnu: Add r-gplots.

[Ricardo Wurmus]

Ludovic Courtès <ludo@gnu.org> writes:

> Leo Famulari <leo@famulari.name> skribis:
>
>> On Thu, Nov 24, 2016 at 05:51:37PM +0100, Ricardo Wurmus wrote:
>>> * gnu/packages/statistics.scm (r-gplots): New variable.
>>
>>> +    (description
>>> +     "This package provides various R programming tools for plotting data,
>>> +including:
>>> +
>>> +@itemize
>>> +@item calculating and plotting locally smoothed summary function
>>> +@item enhanced versions of standard plots
>>> +@item manipulating colors
>>> +@item calculating and plotting two-dimensional data summaries
>>> +@item enhanced regression diagnostic plots
>>> +@item formula-enabled interface to @code{stats::lowess} function
>>> +@item displaying textual data in plots
>>> +@item baloon plots
>>> +@item plotting \"Venn\" diagrams
>>> +@item displaying Open-Office style plots
>>> +@item plotting multiple data on same region, with separate axes
>>> +@item plotting means and confidence intervals
>>> +@item spacing points in an x-y plot so they don't overlap
>>> +@end itemize\n")
>>
>> We had been using @enumerate instead of @itemize to work around
>> <http://bugs.gnu.org/21772>. Do you know if Andy's fix is in our Guile
>> yet?
>
> Yes, it’s in 2.0.12.

Okay, so I don’t need to change it.  Thanks for the review, Leo!  I’ve
pushed this patch series to master
(bd3be46e7fd148eef4b4ba81ee5cd4d159e9c20c).

-- 
Ricardo

GPG: BCA6 89B6 3655 3801 C3C6  2150 197A 5888 235F ACAC
http://elephly.net

[PATCH 17/18] gnu: Add r-rocr.

[Ricardo Wurmus]

* gnu/packages/statistics.scm (r-rocr): New variable.
---
 gnu/packages/statistics.scm | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/gnu/packages/statistics.scm b/gnu/packages/statistics.scm
index f014f16..c2d46f2 100644
--- a/gnu/packages/statistics.scm
+++ b/gnu/packages/statistics.scm
@@ -3685,6 +3685,36 @@ including:
 @end itemize\n")
     (license license:gpl2+)))
 
+(define-public r-rocr
+  (package
+    (name "r-rocr")
+    (version "1.0-7")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "ROCR" version))
+       (sha256
+        (base32
+         "1jay8cm7lgq56i967vm5c2hgaxqkphfpip0gn941li3yhh7p3vz7"))))
+    (properties `((upstream-name . "ROCR")))
+    (build-system r-build-system)
+    (propagated-inputs
+     `(("r-gplots" ,r-gplots)))
+    (home-page "http://rocr.bioinf.mpi-sb.mpg.de/")
+    (synopsis "Visualizing the performance of scoring classifiers")
+    (description
+     "ROCR is a flexible tool for creating cutoff-parameterized 2D performance
+curves by freely combining two from over 25 performance measures (new
+performance measures can be added using a standard interface).  Curves from
+different cross-validation or bootstrapping runs can be averaged by different
+methods, and standard deviations, standard errors or box plots can be used to
+visualize the variability across the runs.  The parameterization can be
+visualized by printing cutoff values at the corresponding curve positions, or
+by coloring the curve according to cutoff.  All components of a performance
+plot can be quickly adjusted using a flexible parameter dispatching
+mechanism.")
+    (license license:gpl2+)))
+
 (define-public r-kernsmooth
   (package
     (name "r-kernsmooth")
-- 
2.10.2

[PATCH 18/18] gnu: Add r-gkmsvm.

[Ricardo Wurmus]

* gnu/packages/bioinformatics.scm (r-gkmsvm): New variable.
---
 gnu/packages/bioinformatics.scm | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index 89a9db1..844bb8e 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -7551,6 +7551,39 @@ bound and non bound genomic regions to accurately identify transcription
 factors bound at the specific regions.")
     (license license:gpl2+)))
 
+(define-public r-gkmsvm
+  (package
+    (name "r-gkmsvm")
+    (version "0.71.0")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "gkmSVM" version))
+       (sha256
+        (base32
+         "1zpxgxmf2nd5j5wn00ps6kfxr8wxh7d1swr1rr4spq7sj5z5z0k0"))))
+    (properties `((upstream-name . "gkmSVM")))
+    (build-system r-build-system)
+    (propagated-inputs
+     `(("r-biocgenerics" ,r-biocgenerics)
+       ("r-biostrings" ,r-biostrings)
+       ("r-genomeinfodb" ,r-genomeinfodb)
+       ("r-genomicranges" ,r-genomicranges)
+       ("r-iranges" ,r-iranges)
+       ("r-kernlab" ,r-kernlab)
+       ("r-rcpp" ,r-rcpp)
+       ("r-rocr" ,r-rocr)
+       ("r-rtracklayer" ,r-rtracklayer)
+       ("r-s4vectors" ,r-s4vectors)
+       ("r-seqinr" ,r-seqinr)))
+    (home-page "http://cran.r-project.org/web/packages/gkmSVM")
+    (synopsis "Gapped-kmer support vector machine")
+    (description
+     "This R package provides tools for training gapped-kmer SVM classifiers
+for DNA and protein sequences.  This package supports several sequence
+kernels, including: gkmSVM, kmer-SVM, mismatch kernel and wildcard kernel.")
+    (license license:gpl2+)))
+
 (define-public emboss
   (package
     (name "emboss")
-- 
2.10.2