diff options
Diffstat (limited to 'gnu/packages/bioconductor.scm')
| -rw-r--r-- | gnu/packages/bioconductor.scm | 446 |
1 files changed, 430 insertions, 16 deletions
diff --git a/gnu/packages/bioconductor.scm b/gnu/packages/bioconductor.scm index 0bcd9a67cd..171736064c 100644 --- a/gnu/packages/bioconductor.scm +++ b/gnu/packages/bioconductor.scm @@ -619,6 +619,32 @@ as provided by UCSC (hg38, Dec. 2013) and stored in Biostrings objects.") "This package exposes an annotation database generated from Ensembl.") (license license:artistic2.0))) +(define-public r-txdb-dmelanogaster-ucsc-dm6-ensgene + (package + (name "r-txdb-dmelanogaster-ucsc-dm6-ensgene") + (version "3.12.0") + (source + (origin + (method url-fetch) + (uri (bioconductor-uri "TxDb.Dmelanogaster.UCSC.dm6.ensGene" + version 'annotation)) + (sha256 + (base32 + "0yij7zyqkmmr13389rs2gfa5anvvw648nnl1kjbsgvyxkggif8q4")))) + (properties + `((upstream-name . "TxDb.Dmelanogaster.UCSC.dm6.ensGene"))) + (build-system r-build-system) + (propagated-inputs + `(("r-annotationdbi" ,r-annotationdbi) + ("r-genomicfeatures" ,r-genomicfeatures))) + (home-page + "https://bioconductor.org/packages/TxDb.Dmelanogaster.UCSC.dm6.ensGene") + (synopsis "Annotation package for TxDb object(s)") + (description + "This package exposes an annotation databases generated from UCSC by +exposing these as TxDb objects.") + (license license:artistic2.0))) + (define-public r-txdb-hsapiens-ucsc-hg19-knowngene (package (name "r-txdb-hsapiens-ucsc-hg19-knowngene") @@ -1087,7 +1113,7 @@ package @code{affy}.") (properties `((upstream-name . "gageData"))) (build-system r-build-system) (home-page "https://bioconductor.org/packages/gageData") - (synopsis "Auxillary data for gage package") + (synopsis "Auxiliary data for the gage package") (description "This is a supportive data package for the software package @code{gage}. However, the data supplied here are also useful for gene set or pathway @@ -1233,6 +1259,39 @@ data. In addition, provides numerous plotting functions for commonly used visualizations.") (license license:artistic2.0))) +(define-public r-delayedarray + (package + (name "r-delayedarray") + (version "0.16.3") + (source (origin + (method url-fetch) + (uri (bioconductor-uri "DelayedArray" version)) + (sha256 + (base32 + "0w1wppy6m2iv41852dscg3y19sq84ahdx3m7c2p2pxjcznmv6hys")))) + (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") @@ -1315,6 +1374,60 @@ naming and share the same rich and consistent \"Vector API\" as much as possible.") (license license:artistic2.0))) +;; This is a CRAN package, but it depends on r-biobase and r-limma from Bioconductor. +(define-public r-absfiltergsea + (package + (name "r-absfiltergsea") + (version "1.5.1") + (source + (origin + (method url-fetch) + (uri (cran-uri "AbsFilterGSEA" version)) + (sha256 + (base32 "15srxkxsvn38kd5frdrwfdf0ad8gskrd0h01wmdf9hglq8fjrp7w")))) + (properties `((upstream-name . "AbsFilterGSEA"))) + (build-system r-build-system) + (propagated-inputs + `(("r-biobase" ,r-biobase) + ("r-deseq" ,r-deseq) + ("r-limma" ,r-limma) + ("r-rcpp" ,r-rcpp) + ("r-rcpparmadillo" ,r-rcpparmadillo))) + (home-page "https://cran.r-project.org/web/packages/AbsFilterGSEA/") + (synopsis "Improved false positive control of gene-permuting with absolute filtering") + (description + "This package provides a function that performs gene-permuting of a gene-set +enrichment analysis (GSEA) calculation with or without the absolute filtering. + Without filtering, users can perform (original) two-tailed or one-tailed +absolute GSEA.") + (license license:gpl2))) + +;; This is a CRAN package, but it depends on r-biobase from Bioconductor. +(define-public r-bisquerna + (package + (name "r-bisquerna") + (version "1.0.4") + (source (origin + (method url-fetch) + (uri (cran-uri "BisqueRNA" version)) + (sha256 + (base32 + "01g34n87ml7n3pck77497ddgbv3rr5p4153ac8ninpgjijlm3jw2")))) + (properties `((upstream-name . "BisqueRNA"))) + (build-system r-build-system) + (propagated-inputs + `(("r-biobase" ,r-biobase) + ("r-limsolve" ,r-limsolve))) + (home-page "https://www.biorxiv.org/content/10.1101/669911v1") + (synopsis "Decomposition of bulk expression with single-cell sequencing") + (description "This package provides tools to accurately estimate cell type +abundances from heterogeneous bulk expression. A reference-based method +utilizes single-cell information to generate a signature matrix and +transformation of bulk expression for accurate regression based estimates. +A marker-based method utilizes known cell-specific marker genes to measure +relative abundances across samples.") + (license license:gpl3))) + ;; This is a CRAN package, but it depends on r-bsgenome-hsapiens-ucsc-hg19 ;; from Bioconductor. (define-public r-deconstructsigs @@ -1559,6 +1672,35 @@ 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") @@ -1591,6 +1733,148 @@ 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") @@ -1764,13 +2048,13 @@ high-throughput sequencing experiments.") (define-public r-genomeinfodb (package (name "r-genomeinfodb") - (version "1.26.4") + (version "1.26.6") (source (origin (method url-fetch) (uri (bioconductor-uri "GenomeInfoDb" version)) (sha256 (base32 - "1sbhdpgabqbi749ixih8nlmq5id7sg8y6kkfj46r00642rc293ys")))) + "1wy4dwiv0pgim975var802z565py4a0nakx6zdvbhry4c0dfczd1")))) (properties `((upstream-name . "GenomeInfoDb"))) (build-system r-build-system) @@ -1791,6 +2075,37 @@ 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") @@ -1871,6 +2186,25 @@ 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") @@ -1975,6 +2309,49 @@ 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") @@ -2110,6 +2487,42 @@ 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") @@ -2318,14 +2731,14 @@ signal in the input, that lead to spurious peaks during peak calling.") (define-public r-diffbind (package (name "r-diffbind") - (version "3.0.14") + (version "3.0.15") (source (origin (method url-fetch) (uri (bioconductor-uri "DiffBind" version)) (sha256 (base32 - "1siabhjd0w7bb6v2gfhsm9j7c7c86z8m6lfsyl8p84h0zhjs2vrw")))) + "06f613s8d9z51njyf839g22gwybx9zs5n6xghwr5j1ad2n4m6qwi")))) (properties `((upstream-name . "DiffBind"))) (build-system r-build-system) (propagated-inputs @@ -2532,14 +2945,14 @@ determining dependencies between variables, code improvement suggestions.") (define-public r-chippeakanno (package (name "r-chippeakanno") - (version "3.24.1") + (version "3.24.2") (source (origin (method url-fetch) (uri (bioconductor-uri "ChIPpeakAnno" version)) (sha256 (base32 - "0qdkwjv8s46d1kmgg2chijv7yzy9sv49kiks18w8x2z89prn15gj")))) + "0l417aygs89wf1j9fjpfjhahzskbpbgcrm8xpx3qm4s0307vfzkw")))) (properties `((upstream-name . "ChIPpeakAnno"))) (build-system r-build-system) (propagated-inputs @@ -2548,6 +2961,7 @@ determining dependencies between variables, code improvement suggestions.") ("r-biomart" ,r-biomart) ("r-biostrings" ,r-biostrings) ("r-dbi" ,r-dbi) + ("r-dplyr" ,r-dplyr) ("r-ensembldb" ,r-ensembldb) ("r-genomeinfodb" ,r-genomeinfodb) ("r-genomicalignments" ,r-genomicalignments) @@ -6521,14 +6935,14 @@ annotations.") (define-public r-rsubread (package (name "r-rsubread") - (version "2.4.2") + (version "2.4.3") (source (origin (method url-fetch) (uri (bioconductor-uri "Rsubread" version)) (sha256 (base32 - "1wczrw5jb69x45hd3rdqqs9vkysdqwlxn9h3kjzn57r4x5q7jrra")))) + "0c4akc89p5467n5rzq9bi7h0h15rbpqpvh7fw42qcj7g2vc41wba")))) (properties `((upstream-name . "Rsubread"))) (build-system r-build-system) (inputs `(("zlib" ,zlib))) @@ -7987,14 +8401,14 @@ data.") (define-public r-universalmotif (package (name "r-universalmotif") - (version "1.8.3") + (version "1.8.4") (source (origin (method url-fetch) (uri (bioconductor-uri "universalmotif" version)) (sha256 (base32 - "1ys2kbayc1rzv8nzi60208yfslm4kzynndfg7vw2n0c30dvzycrc")))) + "0pmi5mp5v0srr482vlkfmkp28bywq969fvv9g5kjl5rxki963zmr")))) (properties `((upstream-name . "universalmotif"))) (build-system r-build-system) @@ -10140,8 +10554,8 @@ experiments, and visualize de influence of the involved factors.") by spectral counts, to discover differentially expressed proteins between two biological conditions. Three tests are available: Poisson GLM regression, quasi-likelihood GLM regression, and the negative binomial of the edgeR -package.The three models admit blocking factors to control for nuissance -variables.To assure a good level of reproducibility a post-test filter is +package. The three models admit blocking factors to control for nuisance +variables. To assure a good level of reproducibility a post-test filter is available, where we may set the minimum effect size considered biologicaly relevant, and the minimum expression of the most abundant condition.") (license license:gpl2))) @@ -10415,14 +10829,14 @@ family of feature/genome hypotheses.") (define-public r-gviz (package (name "r-gviz") - (version "1.34.0") + (version "1.34.1") (source (origin (method url-fetch) (uri (bioconductor-uri "Gviz" version)) (sha256 (base32 - "0v7bz46b91dnrr55ah42ljj1i2xs3090s4w0lw8098pag00p4vh2")))) + "0bmlfz9ri1gkwyl605a2hqi5b8jdpvynrxwghwmrsd657ip6c7n1")))) (properties `((upstream-name . "Gviz"))) (build-system r-build-system) (propagated-inputs @@ -10677,7 +11091,7 @@ the earlier snpMatrix package, allowing for uncertainty in genotypes.") (description "This package implements functions for combinatorial and differential analysis of ChIP-seq data. It includes uni- and multivariate peak-calling, -export to genome browser viewable files, and functi ons for enrichment +export to genome browser viewable files, and functions for enrichment analyses.") (license license:artistic2.0))) |