summaryrefslogtreecommitdiff
path: root/gnu/packages/bioinformatics.scm
diff options
context:
space:
mode:
authorJohn Kehayias <john.kehayias@protonmail.com>2024-01-07 21:27:27 -0500
committerJohn Kehayias <john.kehayias@protonmail.com>2024-01-07 21:27:27 -0500
commitd9dee5ea2f564fa6979ae552fd9bd5ac22f86ecc (patch)
tree8d1b2f89808f9e7ddbeddeab7563f3ab82e533cd /gnu/packages/bioinformatics.scm
parentef4e4c9a2ccc1678182fa6e4409fff13c669fd14 (diff)
parent3de361d9c9d320aefbd43710124d7b07af891de1 (diff)
Merge branch 'master' into mesa-updates
Change-Id: I0b92a95109688a1a4c50572bc25a4e5676e0921e
Diffstat (limited to 'gnu/packages/bioinformatics.scm')
-rw-r--r--gnu/packages/bioinformatics.scm228
1 files changed, 210 insertions, 18 deletions
diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index 41cac296fe..7dbebcf3da 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -130,6 +130,7 @@
#:use-module (gnu packages pdf)
#:use-module (gnu packages perl)
#:use-module (gnu packages perl-check)
+ #:use-module (gnu packages perl-web)
#:use-module (gnu packages pkg-config)
#:use-module (gnu packages popt)
#:use-module (gnu packages protobuf)
@@ -574,6 +575,30 @@ BED, GFF/GTF, VCF.")
whole-genome bisulfite sequencing (WGBS) reads from directional protocol.")
(license license:asl2.0)))
+(define-public bustools
+ (package
+ (name "bustools")
+ (version "0.43.2")
+ (source (origin
+ (method git-fetch)
+ (uri (git-reference
+ (url "https://github.com/BUStools/bustools")
+ (commit (string-append "v" version))))
+ (file-name (git-file-name name version))
+ (sha256
+ (base32
+ "068kjlc4d528269nl5mc3j8h2c95r1v545d3fi1iw1ckg8rba0hg"))))
+ (build-system cmake-build-system)
+ (arguments (list #:tests? #f)) ;no test target
+ (inputs (list zlib))
+ (home-page "https://bustools.github.io")
+ (synopsis "Tools for working with BUS files")
+ (description "bustools is a program for manipulating BUS files for single
+cell RNA-Seq datasets. It can be used to error correct barcodes, collapse
+UMIs, produce gene count or transcript compatibility count matrices, and is useful
+for many other tasks.")
+ (license license:bsd-2)))
+
(define-public cellsnp-lite
;; Last release is from November 2021 and does not contain fixes.
(let ((commit "0885d746b0b1ea65c8ef92f8943ca7669ca9734a")
@@ -2365,6 +2390,40 @@ sequencing data and the end result are tables of UMI-unique DamID and CEL-Seq
counts.")
(license license:expat)))
+(define-public python-snaptools
+ (package
+ (name "python-snaptools")
+ (version "1.4.8")
+ (source
+ (origin
+ (method url-fetch)
+ (uri (pypi-uri "snaptools" version))
+ (sha256
+ (base32
+ "1s5373g5jjbshh3q39zy7dlxr7nda6ksxq9d1gw46h82c4fsmfbn"))))
+ (build-system pyproject-build-system)
+ (propagated-inputs
+ (list python-future
+ python-h5py
+ python-louvain
+ python-numpy
+ python-pybedtools
+ python-pysam))
+ (home-page "https://github.com/r3fang/SnapTools")
+ (synopsis "Tools for processing snap files" )
+ (description
+ "@code{SnapTools} can operate on snap files the following types of
+operations:
+
+@itemize
+@item index the reference genome before alignment;
+@item align reads to the corresponding reference genome;
+@item pre-process by convert pair-end reads into fragments, checking the
+ mapping quality score, alingment and filtration;
+@item create the cell-by-bin matrix.
+@end itemize")
+ (license license:asl2.0)))
+
(define-public python-bioframe
(package
(name "python-bioframe")
@@ -4999,6 +5058,126 @@ be of arbitrary length. Repeats with pattern size in the range from 1 to 2000
bases are detected.")
(license license:agpl3+)))
+(define-public trinityrnaseq
+ (package
+ (name "trinityrnaseq")
+ (version "2.13.2")
+ (source (origin
+ (method git-fetch)
+ (uri (git-reference
+ (url "https://github.com/trinityrnaseq/trinityrnaseq.git")
+ (commit (string-append "Trinity-v" version))
+ (recursive? #true)))
+ (file-name (git-file-name name version))
+ (sha256
+ (base32
+ "1qszrxqbx4q5pavpgm4rkrh1z1v1mf7qx83vv3fnlqdmncnsf1gv"))))
+ (build-system gnu-build-system)
+ (arguments
+ (list
+ #:test-target "test"
+ #:modules
+ '((guix build gnu-build-system)
+ (guix build utils)
+ (ice-9 match)
+ (srfi srfi-1))
+ #:make-flags
+ #~(list (string-append "CC=" #$(cc-for-target)))
+ #:phases
+ #~(modify-phases %standard-phases
+ (replace 'configure
+ (lambda _
+ (setenv "SHELL" (which "sh"))
+ (setenv "CONFIG_SHELL" (which "sh"))
+ ;; Do not require version.h, which triggers a local build of a
+ ;; vendored htslib.
+ (substitute* "trinity-plugins/bamsifter/Makefile"
+ (("sift_bam_max_cov.cpp htslib/version.h")
+ "sift_bam_max_cov.cpp"))))
+ (add-after 'build 'build-plugins
+ (lambda _
+ ;; Run this in the subdirectory to avoid running the
+ ;; tests right here.
+ (with-directory-excursion "trinity-plugins"
+ (invoke "make" "plugins"))))
+ ;; The install script uses rsync, provides no overrides for the
+ ;; default location at /usr/local/bin, and patching it would change
+ ;; all lines that do something.
+ (replace 'install
+ (lambda* (#:key inputs #:allow-other-keys)
+ (let ((share (string-append #$output "/share/trinity/"))
+ (bin (string-append #$output "/bin/")))
+ (mkdir-p bin)
+ (copy-recursively "." share)
+ (delete-file (string-append share "/Chrysalis/build/CMakeFiles/CMakeOutput.log"))
+ (delete-file (string-append share "/Inchworm/build/CMakeFiles/CMakeOutput.log"))
+
+ (wrap-program (string-append share "Trinity")
+ `("R_LIBS_SITE" ":" = (,(getenv "R_LIBS_SITE")))
+ `("PERL5LIB" ":" = (,(getenv "PERL5LIB")))
+ `("PYTHONPATH" ":" = (,(getenv "GUIX_PYTHONPATH")))
+ `("PATH" ":" =
+ ,(cons (string-append share "/trinity-plugins/BIN")
+ (filter-map (match-lambda
+ ((name . dir)
+ (string-append dir "/bin")))
+ inputs))))
+ (symlink (string-append share "Trinity")
+ (string-append bin "Trinity"))))))))
+ (inputs
+ (list blast+
+ bowtie
+ fastqc
+ hisat
+ htslib
+ icedtea-8
+ jellyfish
+ kallisto
+ multiqc
+ perl
+ perl-uri-escape
+ python-numpy
+ python-wrapper
+ r-ape
+ r-argparse
+ r-biobase
+ r-ctc
+ r-deseq2
+ r-edger
+ r-fastcluster
+ r-glimma
+ r-goplot
+ r-goseq
+ r-gplots
+ r-minimal
+ r-qvalue
+ r-rots
+ r-sm
+ r-tidyverse
+ rsem
+ salmon
+ samtools
+ sra-tools
+ star
+ zlib))
+ (propagated-inputs
+ (list coreutils
+ gzip
+ which))
+ (native-inputs (list cmake))
+ (home-page "https://github.com/trinityrnaseq/trinityrnaseq/wiki")
+ (synopsis "Trinity RNA-Seq de novo transcriptome assembly")
+ (description "Trinity assembles transcript sequences from Illumina RNA-Seq
+data. Trinity represents a novel method for the efficient and robust de novo
+reconstruction of transcriptomes from RNA-seq data. Trinity combines three
+independent software modules: Inchworm, Chrysalis, and Butterfly, applied
+sequentially to process large volumes of RNA-seq reads. Trinity partitions
+the sequence data into many individual de Bruijn graphs, each representing the
+transcriptional complexity at a given gene or locus, and then processes each
+graph independently to extract full-length splicing isoforms and to tease
+apart transcripts derived from paralogous genes.")
+ (license license:bsd-3)))
+
(define-public repeat-masker
(package
(name "repeat-masker")
@@ -16790,7 +16969,16 @@ the HiCExplorer and pyGenomeTracks packages.")
(file-name (git-file-name name version))
(sha256
(base32
- "1yavgxry38g326z10bclvdf8glmma05fxj5m73h15m1r2l9xmw3v"))))
+ "1yavgxry38g326z10bclvdf8glmma05fxj5m73h15m1r2l9xmw3v"))
+ (modules '((guix build utils)))
+ ;; setup.py is malformed. The requirements are defined using a catchall
+ ;; pattern for the patch version number. This has been fixed in version
+ ;; 3.7.3, but we cannot upgrade to this version yet, since some Guix
+ ;; packages are not new enough. (See upstream commit
+ ;; 4845c715ec7b105e938d0c2426e27d0181690bfe for the fix).
+ (snippet '(substitute* "setup.py"
+ (("\\.\\*")
+ "")))))
(build-system pyproject-build-system)
(arguments
(list
@@ -20512,24 +20700,28 @@ aligner.")
(uri (pypi-uri "scvelo" version))
(sha256
(base32 "0h5ha1459ljs0qgpnlfsw592i8dxqn6p9bl08l1ikpwk36baxb7z"))))
- (build-system python-build-system)
+ (build-system pyproject-build-system)
(arguments
- `(#:phases
- (modify-phases %standard-phases
- ;; Numba needs a writable dir to cache functions.
- (add-before 'check 'set-numba-cache-dir
- (lambda _
- (setenv "NUMBA_CACHE_DIR" "/tmp")))
- (replace 'check
- (lambda* (#:key outputs tests? #:allow-other-keys)
- (when tests?
- ;; The discovered test file names must match the names of the
- ;; compiled files, so we cannot run the tests from
- ;; /tmp/guix-build-*.
- (with-directory-excursion
- (string-append (assoc-ref outputs "out")
- "/lib/python3.10/site-packages/scvelo/core/tests/")
- (invoke "pytest" "-v"))))))))
+ (list
+ #:test-flags
+ ;; XXX: these two tests fail for unknown reasons
+ '(list "-k" "not test_perfect_fit and not test_perfect_fit_2d")
+ #:phases
+ #~(modify-phases %standard-phases
+ ;; Numba needs a writable dir to cache functions.
+ (add-before 'check 'set-numba-cache-dir
+ (lambda _
+ (setenv "NUMBA_CACHE_DIR" "/tmp")))
+ (replace 'check
+ (lambda* (#:key tests? test-flags #:allow-other-keys)
+ (when tests?
+ ;; The discovered test file names must match the names of the
+ ;; compiled files, so we cannot run the tests from
+ ;; /tmp/guix-build-*.
+ (with-directory-excursion
+ (string-append #$output
+ "/lib/python3.10/site-packages/scvelo/core/tests/")
+ (apply invoke "pytest" "-v" test-flags))))))))
(propagated-inputs
(list python-anndata
python-hnswlib