Toggle diff (517 lines)
diff --git a/gnu/packages/bioconductor.scm b/gnu/packages/bioconductor.scm
index 6530a31fb5..d9e938db5c 100644
--- a/gnu/packages/bioconductor.scm
+++ b/gnu/packages/bioconductor.scm
@@ -1212,6 +1212,39 @@ data. In addition, provides numerous plotting functions for commonly
used visualizations.")
(license license:artistic2.0)))
+(define-public r-delayedarray
+ (package
+ (name "r-delayedarray")
+ (version "0.16.2")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "DelayedArray" version))
+ (sha256
+ (base32
+ "09lpj951v1afxkrnjvnhzp4qgklq23ykdwlny7k1lyfcdy9q6wm0"))))
+ (properties
+ `((upstream-name . "DelayedArray")))
+ (build-system r-build-system)
+ (propagated-inputs
+ `(("r-biocgenerics" ,r-biocgenerics)
+ ("r-s4vectors" ,r-s4vectors)
+ ("r-iranges" ,r-iranges)
+ ("r-matrix" ,r-matrix)
+ ("r-matrixgenerics" ,r-matrixgenerics)))
+ (native-inputs
+ `(("r-knitr" ,r-knitr)))
+ (home-page "https://bioconductor.org/packages/DelayedArray")
+ (synopsis "Delayed operations on array-like objects")
+ (description
+ "Wrapping an array-like object (typically an on-disk object) in a
+@code{DelayedArray} object allows one to perform common array operations on it
+without loading the object in memory. In order to reduce memory usage and
+optimize performance, operations on the object are either delayed or executed
+using a block processing mechanism. Note that this also works on in-memory
+array-like objects like @code{DataFrame} objects (typically with Rle columns),
+@code{Matrix} objects, and ordinary arrays and data frames.")
+ (license license:artistic2.0)))
+
(define-public r-bluster
(package
(name "r-bluster")
@@ -1592,6 +1625,35 @@ structure.")
microarrays.")
(license license:artistic2.0)))
+(define-public r-annotationdbi
+ (package
+ (name "r-annotationdbi")
+ (version "1.52.0")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "AnnotationDbi" version))
+ (sha256
+ (base32
+ "0zqxgh3nx6y8ry12s2vss2f4axz5vpqxha1y4ifhhcx4zhpzsglr"))))
+ (properties
+ `((upstream-name . "AnnotationDbi")))
+ (build-system r-build-system)
+ (propagated-inputs
+ `(("r-biobase" ,r-biobase)
+ ("r-biocgenerics" ,r-biocgenerics)
+ ("r-dbi" ,r-dbi)
+ ("r-iranges" ,r-iranges)
+ ("r-rsqlite" ,r-rsqlite)
+ ("r-s4vectors" ,r-s4vectors)))
+ (native-inputs
+ `(("r-knitr" ,r-knitr)))
+ (home-page "https://bioconductor.org/packages/AnnotationDbi")
+ (synopsis "Annotation database interface")
+ (description
+ "This package provides user interface and database connection code for
+annotation data packages using SQLite data storage.")
+ (license license:artistic2.0)))
+
(define-public r-annotationforge
(package
(name "r-annotationforge")
@@ -1624,6 +1686,148 @@ microarrays.")
databases. Packages produced are intended to be used with AnnotationDbi.")
(license license:artistic2.0)))
+(define-public r-biobase
+ (package
+ (name "r-biobase")
+ (version "2.50.0")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "Biobase" version))
+ (sha256
+ (base32
+ "11kgc4flywlm3i18603558l8ksv91c24vkc5fnnbcd375i2dhhd4"))))
+ (properties
+ `((upstream-name . "Biobase")))
+ (build-system r-build-system)
+ (propagated-inputs
+ `(("r-biocgenerics" ,r-biocgenerics)))
+ (home-page "https://bioconductor.org/packages/Biobase")
+ (synopsis "Base functions for Bioconductor")
+ (description
+ "This package provides functions that are needed by many other packages
+on Bioconductor or which replace R functions.")
+ (license license:artistic2.0)))
+
+(define-public r-biomart
+ (package
+ (name "r-biomart")
+ (version "2.46.3")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "biomaRt" version))
+ (sha256
+ (base32
+ "0gwmd0ykpv0gyh34c56g5m12lil20fvig49f3ih1jxrxf3q4wmq7"))))
+ (properties
+ `((upstream-name . "biomaRt")))
+ (build-system r-build-system)
+ (propagated-inputs
+ `(("r-annotationdbi" ,r-annotationdbi)
+ ("r-biocfilecache" ,r-biocfilecache)
+ ("r-httr" ,r-httr)
+ ("r-openssl" ,r-openssl)
+ ("r-progress" ,r-progress)
+ ("r-rappdirs" ,r-rappdirs)
+ ("r-stringr" ,r-stringr)
+ ("r-xml" ,r-xml)
+ ("r-xml2" ,r-xml2)))
+ (native-inputs
+ `(("r-knitr" ,r-knitr)))
+ (home-page "https://bioconductor.org/packages/biomaRt")
+ (synopsis "Interface to BioMart databases")
+ (description
+ "biomaRt provides an interface to a growing collection of databases
+implementing the @url{BioMart software suite, http://www.biomart.org}. The
+package enables retrieval of large amounts of data in a uniform way without
+the need to know the underlying database schemas or write complex SQL queries.
+Examples of BioMart databases are Ensembl, COSMIC, Uniprot, HGNC, Gramene,
+Wormbase and dbSNP mapped to Ensembl. These major databases give biomaRt
+users direct access to a diverse set of data and enable a wide range of
+powerful online queries from gene annotation to database mining.")
+ (license license:artistic2.0)))
+
+(define-public r-biocparallel
+ (package
+ (name "r-biocparallel")
+ (version "1.24.1")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "BiocParallel" version))
+ (sha256
+ (base32
+ "1iryicvmcagcrj29kp49mqhiq2kn72j4idj380hi9illmdrg9ism"))))
+ (properties
+ `((upstream-name . "BiocParallel")))
+ (build-system r-build-system)
+ (arguments
+ `(#:phases
+ (modify-phases %standard-phases
+ (add-after 'unpack 'make-reproducible
+ (lambda _
+ ;; Remove generated documentation.
+ (for-each delete-file
+ '("inst/doc/BiocParallel_BatchtoolsParam.pdf"
+ "inst/doc/Introduction_To_BiocParallel.pdf"
+ "inst/doc/Errors_Logs_And_Debugging.pdf"
+ "inst/doc/BiocParallel_BatchtoolsParam.R"
+ "inst/doc/Introduction_To_BiocParallel.R"
+ "inst/doc/Errors_Logs_And_Debugging.R"))
+
+ ;; Remove time-dependent macro
+ (substitute* '("inst/doc/BiocParallel_BatchtoolsParam.Rnw"
+ "inst/doc/Introduction_To_BiocParallel.Rnw"
+ "inst/doc/Errors_Logs_And_Debugging.Rnw"
+ "vignettes/BiocParallel_BatchtoolsParam.Rnw"
+ "vignettes/Introduction_To_BiocParallel.Rnw"
+ "vignettes/Errors_Logs_And_Debugging.Rnw")
+ (("\\today") "later"))
+
+ ;; Initialize the random number generator seed when building.
+ (substitute* "R/internal_rng_stream.R"
+ (("\"L'Ecuyer-CMRG\"\\)" m)
+ (string-append
+ m "; if (!is.na(Sys.getenv(\"SOURCE_DATE_EPOCH\"))) {set.seed(100)}\n"))))))))
+ (propagated-inputs
+ `(("r-futile-logger" ,r-futile-logger)
+ ("r-snow" ,r-snow)
+ ("r-bh" ,r-bh)))
+ (native-inputs
+ `(("r-knitr" ,r-knitr)))
+ (home-page "https://bioconductor.org/packages/BiocParallel")
+ (synopsis "Bioconductor facilities for parallel evaluation")
+ (description
+ "This package provides modified versions and novel implementation of
+functions for parallel evaluation, tailored to use with Bioconductor
+objects.")
+ (license (list license:gpl2+ license:gpl3+))))
+
+(define-public r-biostrings
+ (package
+ (name "r-biostrings")
+ (version "2.58.0")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "Biostrings" version))
+ (sha256
+ (base32
+ "1rbqhs73mhfr1gi0rx28jiyan7i3hb45ai3jpl1656fnrhgjfxq5"))))
+ (properties
+ `((upstream-name . "Biostrings")))
+ (build-system r-build-system)
+ (propagated-inputs
+ `(("r-biocgenerics" ,r-biocgenerics)
+ ("r-crayon" ,r-crayon)
+ ("r-iranges" ,r-iranges)
+ ("r-s4vectors" ,r-s4vectors)
+ ("r-xvector" ,r-xvector)))
+ (home-page "https://bioconductor.org/packages/Biostrings")
+ (synopsis "String objects and algorithms for biological sequences")
+ (description
+ "This package provides memory efficient string containers, string
+matching algorithms, and other utilities, for fast manipulation of large
+biological sequences or sets of sequences.")
+ (license license:artistic2.0)))
+
(define-public r-category
(package
(name "r-category")
@@ -1824,6 +2028,37 @@ translation between different chromosome sequence naming conventions (e.g.,
names in their natural, rather than lexicographic, order.")
(license license:artistic2.0)))
+(define-public r-genomicranges
+ (package
+ (name "r-genomicranges")
+ (version "1.42.0")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "GenomicRanges" version))
+ (sha256
+ (base32
+ "0j4py5g6pdj35xhlaqhxxhg55j9l4mcdk3yck4dgyavv5f2dh24i"))))
+ (properties
+ `((upstream-name . "GenomicRanges")))
+ (build-system r-build-system)
+ (propagated-inputs
+ `(("r-biocgenerics" ,r-biocgenerics)
+ ("r-genomeinfodb" ,r-genomeinfodb)
+ ("r-iranges" ,r-iranges)
+ ("r-s4vectors" ,r-s4vectors)
+ ("r-xvector" ,r-xvector)))
+ (native-inputs
+ `(("r-knitr" ,r-knitr)))
+ (home-page "https://bioconductor.org/packages/GenomicRanges")
+ (synopsis "Representation and manipulation of genomic intervals")
+ (description
+ "This package provides tools to efficiently represent and manipulate
+genomic annotations and alignments is playing a central role when it comes to
+analyzing high-throughput sequencing data (a.k.a. NGS data). The
+GenomicRanges package defines general purpose containers for storing and
+manipulating genomic intervals and variables defined along a genome.")
+ (license license:artistic2.0)))
+
(define-public r-gostats
(package
(name "r-gostats")
@@ -1904,6 +2139,25 @@ Enrichment Analysis} (GSEA).")
the Human Protein Atlas project.")
(license license:artistic2.0)))
+(define-public r-limma
+ (package
+ (name "r-limma")
+ (version "3.46.0")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "limma" version))
+ (sha256
+ (base32
+ "1xxv493q1kip9bjfv7v7k5dnq7hz7gvl80i983v4mvwavhgnbxfz"))))
+ (build-system r-build-system)
+ (home-page "http://bioinf.wehi.edu.au/limma")
+ (synopsis "Package for linear models for microarray and RNA-seq data")
+ (description "This package can be used for the analysis of gene expression
+studies, especially the use of linear models for analysing designed experiments
+and the assessment of differential expression. The analysis methods apply to
+different technologies, including microarrays, RNA-seq, and quantitative PCR.")
+ (license license:gpl2+)))
+
(define-public r-rbgl
(package
(name "r-rbgl")
@@ -2008,6 +2262,49 @@ reports together for a particular project that can be viewed in a web
browser.")
(license license:artistic2.0)))
+(define-public r-rsamtools
+ (package
+ (name "r-rsamtools")
+ (version "2.6.0")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "Rsamtools" version))
+ (sha256
+ (base32
+ "040pggkwglc6wy90qnc7xcdnaj0v3iqlykvvsl74241409qly554"))))
+ (properties
+ `((upstream-name . "Rsamtools")))
+ (build-system r-build-system)
+ (arguments
+ `(#:phases
+ (modify-phases %standard-phases
+ (add-after 'unpack 'use-system-zlib
+ (lambda _
+ (substitute* "DESCRIPTION"
+ (("zlibbioc, ") ""))
+ (substitute* "NAMESPACE"
+ (("import\\(zlibbioc\\)") ""))
+ #t)))))
+ (propagated-inputs
+ `(("r-biocgenerics" ,r-biocgenerics)
+ ("r-biocparallel" ,r-biocparallel)
+ ("r-biostrings" ,r-biostrings)
+ ("r-bitops" ,r-bitops)
+ ("r-genomeinfodb" ,r-genomeinfodb)
+ ("r-genomicranges" ,r-genomicranges)
+ ("r-iranges" ,r-iranges)
+ ("r-rhtslib" ,r-rhtslib)
+ ("r-s4vectors" ,r-s4vectors)
+ ("r-xvector" ,r-xvector)))
+ (home-page "https://bioconductor.org/packages/release/bioc/html/Rsamtools.html")
+ (synopsis "Interface to samtools, bcftools, and tabix")
+ (description
+ "This package provides an interface to the @code{samtools},
+@code{bcftools}, and @code{tabix} utilities for manipulating SAM (Sequence
+Alignment / Map), FASTA, binary variant call (BCF) and compressed indexed
+tab-delimited (tabix) files.")
+ (license license:expat)))
+
(define-public r-shortread
(package
(name "r-shortread")
@@ -2143,6 +2440,42 @@ annotation infrastructure.")
coding changes and predict coding outcomes.")
(license license:artistic2.0)))
+(define-public r-xvector
+ (package
+ (name "r-xvector")
+ (version "0.30.0")
+ (source (origin
+ (method url-fetch)
+ (uri (bioconductor-uri "XVector" version))
+ (sha256
+ (base32
+ "1pqljikg4f6jb7wgm5537zwgq5b013nyz1agjrwfq2cljb0ym6lq"))))
+ (properties
+ `((upstream-name . "XVector")))
+ (build-system r-build-system)
+ (arguments
+ `(#:phases
+ (modify-phases %standard-phases
+ (add-after 'unpack 'use-system-zlib
+ (lambda _
+ (substitute* "DESCRIPTION"
+ (("zlibbioc, ") ""))
+ (substitute* "NAMESPACE"
+ (("import\\(zlibbioc\\)") ""))
+ #t)))))
+ (inputs
+ `(("zlib" ,zlib)))
+ (propagated-inputs
+ `(("r-biocgenerics" ,r-biocgenerics)
+ ("r-iranges" ,r-iranges)
+ ("r-s4vectors" ,r-s4vectors)))
+ (home-page "https://bioconductor.org/packages/XVector")
+ (synopsis "Representation and manpulation of external sequences")
+ (description
+ "This package provides memory efficient S4 classes for storing sequences
+\"externally\" (behind an R external pointer, or on disk).")
+ (license license:artistic2.0)))
+
(define-public r-geneplotter
(package
(name "r-geneplotter")
diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index eb466868d1..51301a9603 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -7695,311 +7695,6 @@ including VCF header and contents in RDF and JSON.")
(home-page "https://github.com/vcflib/bio-vcf")
(license license:expat)))
-(define-public r-limma
- (package
- (name "r-limma")
- (version "3.46.0")
- (source (origin
- (method url-fetch)
- (uri (bioconductor-uri "limma" version))
- (sha256
- (base32
- "1xxv493q1kip9bjfv7v7k5dnq7hz7gvl80i983v4mvwavhgnbxfz"))))
- (build-system r-build-system)
- (home-page "http://bioinf.wehi.edu.au/limma")
- (synopsis "Package for linear models for microarray and RNA-seq data")
- (description "This package can be used for the analysis of gene expression
-studies, especially the use of linear models for analysing designed experiments
-and the assessment of differential expression. The analysis methods apply to
-different technologies, including microarrays, RNA-seq, and quantitative PCR.")
- (license license:gpl2+)))
-
-(define-public r-xvector
- (package
- (name "r-xvector")
- (version "0.30.0")
- (source (origin
- (method url-fetch)
- (uri (bioconductor-uri "XVector" version))
- (sha256
- (base32
- "1pqljikg4f6jb7wgm5537zwgq5b013nyz1agjrwfq2cljb0ym6lq"))))
- (properties
- `((upstream-name . "XVector")))
- (build-system r-build-system)
- (arguments
- `(#:phases
- (modify-phases %standard-phases
- (add-after 'unpack 'use-system-zlib
- (lambda _
- (substitute* "DESCRIPTION"
- (("zlibbioc, ") ""))
- (substitute* "NAMESPACE"
- (("import\\(zlibbioc\\)") ""))
- #t)))))
- (inputs
- `(("zlib" ,zlib)))
- (propagated-inputs
- `(("r-biocgenerics" ,r-biocgenerics)
- ("r-iranges" ,r-iranges)
- ("r-s4vectors" ,r-s4vectors)))
- (home-page "https://bioconductor.org/packages/XVector")
- (synopsis "Representation and manpulation of external sequences")
- (description
- "This package provides memory efficient S4 classes for storing sequences
-\"externally\" (behind an R external pointer, or on disk).")
- (license license:artistic2.0)))
-
-(define-public r-genomicranges
- (package
- (name "r-genomicranges")
- (version "1.42.0")
- (source (origin
- (method url-fetch)
- (uri (bioconductor-uri "GenomicRanges" version))
- (sha256
- (base32
- "0j4py5g6pdj35xhlaqhxxhg55j9l4mcdk3yck4dgyavv5f2dh24i"))))
- (properties
- `((upstream-name . "GenomicRanges")))
- (build-system r-build-system)
- (propagated-inputs
- `(("r-biocgenerics" ,r-biocgenerics)
- ("r-genomeinfodb" ,r-genomeinfodb)
- ("r-iranges" ,r-iranges)
- ("r-s4vectors" ,r-s4vectors)
- ("r-xvector" ,r-xvector)))
- (native-inputs
- `(("r-knitr" ,r-knitr)))
- (home-page "https://bioconductor.org/packages/GenomicRanges")
- (synopsis "Representation and manipulation of genomic intervals")
- (description
- "This package provides tools to efficiently represent and manipulate
-genomic annotations and alignments is playing a central role when it comes to
-analyzing high-throughput sequencing data (a.k.a. NGS data). The
-GenomicRanges package defines general purpose containers for storing and
-manipulating genomic intervals and variables defined along a genome.")
- (license license:artistic2.0)))
-
-(define-public r-biobase
- (package
- (name "r-biobase")
- (version "2.50.0")
- (source (origin
- (method url-fetch)
- (uri (bioconductor-uri "Biobase" version))
- (sha256
- (base32
- "11kgc4flywlm3i18603558l8ksv91c24vkc5fnnbcd375i2dhhd4"))))
- (properties
- `((upstream-name . "Biobase")))
- (build-system r-build-system)
- (propagated-inputs
- `(("r-biocgenerics" ,r-biocgenerics)))
- (home-page "https://bioconductor.org/packages/Biobase")
- (synopsis "Base functions for Bioconductor")
- (description
- "This package provides functions that are needed by many other packages
-on Bioconductor or which replace R functions.")
- (license license:artistic2.0)))
-
-(define-public r-annotationdbi
- (package
- (name "r-annotationdbi")
- (version "1.52.0")
- (source (origin
- (method url-fetch)
- (uri (bioconductor-uri "AnnotationDbi" version))
- (sha256
- (base32
- "0zqxgh3nx6y8ry12s2vss2f4axz5vpqxha1y4ifhhcx4zhpzsglr"))))
- (properties
- `((upstream-name . "AnnotationDbi")))
- (build-system r-build-system)
- (propagated-inputs
-
