feat: Add wrappers in bio/reference for Ensembl Genomes

bio/reference/ensembl-genomes-annotation/environment.linux-64.pin.txt

@@ -0,0 +1,26 @@
+# This file may be used to create an environment using:
+# $ conda create --name <env> --file <this file>
+# platform: linux-64
+# created-by: conda 25.3.1
+@EXPLICIT
+https://conda.anaconda.org/conda-forge/linux-64/_libgcc_mutex-0.1-conda_forge.tar.bz2#d7c89558ba9fa0495403155b64376d81
+https://conda.anaconda.org/conda-forge/noarch/ca-certificates-2025.11.12-hbd8a1cb_0.conda#f0991f0f84902f6b6009b4d2350a83aa
+https://conda.anaconda.org/conda-forge/linux-64/libgomp-15.2.0-he0feb66_16.conda#26c46f90d0e727e95c6c9498a33a09f3
+https://conda.anaconda.org/conda-forge/linux-64/_openmp_mutex-4.5-2_gnu.tar.bz2#73aaf86a425cc6e73fcf236a5a46396d
+https://conda.anaconda.org/conda-forge/linux-64/libgcc-15.2.0-he0feb66_16.conda#6d0363467e6ed84f11435eb309f2ff06
+https://conda.anaconda.org/conda-forge/linux-64/c-ares-1.34.6-hb03c661_0.conda#920bb03579f15389b9e512095ad995b7
+https://conda.anaconda.org/conda-forge/linux-64/keyutils-1.6.3-hb9d3cd8_0.conda#b38117a3c920364aff79f870c984b4a3
+https://conda.anaconda.org/conda-forge/linux-64/libgcc-ng-15.2.0-h69a702a_16.conda#5a68259fac2da8f2ee6f7bfe49c9eb8b
+https://conda.anaconda.org/conda-forge/linux-64/libstdcxx-15.2.0-h934c35e_16.conda#68f68355000ec3f1d6f26ea13e8f525f
+https://conda.anaconda.org/conda-forge/linux-64/libzlib-1.3.1-hb9d3cd8_2.conda#edb0dca6bc32e4f4789199455a1dbeb8
+https://conda.anaconda.org/conda-forge/linux-64/ncurses-6.5-h2d0b736_3.conda#47e340acb35de30501a76c7c799c41d7
+https://conda.anaconda.org/conda-forge/linux-64/openssl-3.6.0-h26f9b46_0.conda#9ee58d5c534af06558933af3c845a780
+https://conda.anaconda.org/conda-forge/linux-64/libedit-3.1.20250104-pl5321h7949ede_0.conda#c277e0a4d549b03ac1e9d6cbbe3d017b
+https://conda.anaconda.org/conda-forge/linux-64/libev-4.33-hd590300_2.conda#172bf1cd1ff8629f2b1179945ed45055
+https://conda.anaconda.org/conda-forge/linux-64/libssh2-1.11.1-hcf80075_0.conda#eecce068c7e4eddeb169591baac20ac4
+https://conda.anaconda.org/conda-forge/linux-64/libstdcxx-ng-15.2.0-hdf11a46_16.conda#1b3152694d236cf233b76b8c56bf0eae
+https://conda.anaconda.org/conda-forge/linux-64/zstd-1.5.7-hb78ec9c_6.conda#4a13eeac0b5c8e5b8ab496e6c4ddd829
+https://conda.anaconda.org/conda-forge/linux-64/krb5-1.21.3-h659f571_0.conda#3f43953b7d3fb3aaa1d0d0723d91e368
+https://conda.anaconda.org/conda-forge/linux-64/libnghttp2-1.67.0-had1ee68_0.conda#b499ce4b026493a13774bcf0f4c33849
+https://conda.anaconda.org/conda-forge/linux-64/libcurl-8.17.0-h4e3cde8_1.conda#117499f93e892ea1e57fdca16c2e8351
+https://conda.anaconda.org/conda-forge/linux-64/curl-8.17.0-h4e3cde8_1.conda#cb3648a22e7d5010fc9edae69bb2eab8

bio/reference/ensembl-genomes-annotation/environment.yaml

@@ -0,0 +1,5 @@
+channels:
+  - conda-forge
+  - nodefaults
+dependencies:
+  - curl

bio/reference/ensembl-genomes-annotation/meta.yaml

@@ -0,0 +1,13 @@
+name: ensembl-genomes-annotation
+description: |
+  Download annotation of genomic sites (e.g. transcripts) for non-vertebrate
+  species from Ensembl Genomes FTP servers, and store them in a single .gtf or
+  .gff3 file. Adapted from ``ensembl-annotation`` wrapper.
+authors:
+  - Brandon Seah
+  - Johannes Köster
+url: https://ensemblgenomes.org/
+output:
+  - Ensembl Genomes GTF or GFF3 annotation file
+params:
+  - url: URL from where to download cache data (optional; by default is ``https://ftp.ebi.ac.uk/ensemblgenomes/pub``)

bio/reference/ensembl-genomes-annotation/test/Snakefile

@@ -0,0 +1,28 @@
+rule get_annotation:
+    output:
+        "refs/ensembl-genomes-annotation.gtf",
+    params:
+        species="schizosaccharomyces_pombe",
+        division="fungi",
+        assembly="ASM294v2",
+        release="62",
+    log:
+        "logs/get_annotation.log",
+    cache: "omit-software"  # save space and time with between workflow caching (see docs)
+    wrapper:
+        "master/bio/reference/ensembl-genomes-annotation"
+
+
+rule get_annotation_gz:
+    output:
+        "refs/ensembl-genomes-annotation.gtf.gz",
+    params:
+        species="schizosaccharomyces_pombe",
+        division="fungi",
+        assembly="ASM294v2",
+        release="62",
+    log:
+        "logs/get_annotation.log",
+    cache: "omit-software"  # save space and time with between workflow caching (see docs)
+    wrapper:
+        "master/bio/reference/ensembl-genomes-annotation"

bio/reference/ensembl-genomes-annotation/wrapper.py

@@ -0,0 +1,67 @@
+__author__ = "Brandon Seah, Johannes Köster"
+__copyright__ = "Copyright 2025, Brandon Seah, Johannes Köster"
+__email__ = "brandon_seah@tll.org.sg"
+__license__ = "MIT"
+
+import subprocess as sp
+import sys
+
+from snakemake.shell import shell
+from snakemake.logging import logger
+from pathlib import Path
+
+species = snakemake.params.species.lower()
+release = int(snakemake.params.release)
+division = snakemake.params.division.lower()
+assembly = snakemake.params.assembly
+
+out_fmt = Path(snakemake.output[0]).suffixes
+out_gz = (out_fmt.pop() and True) if out_fmt[-1] == ".gz" else False
+out_fmt = out_fmt.pop().lstrip(".")
+
+available_divisions = ["fungi", "metazoa", "plants", "protists"]
+if division not in available_divisions:
+    raise ValueError(
+        f"Invalid division. Division must be one of: {', '.join(available_divisions)}"
+    )
+
+# TODO: bacteria, fungi/ascomycota folders are further subdivided
+# TODO: chromosome file names may not follow standard naming convention, e.g.
+# Plasmodium_falciparum.GCA000002765v3.dna.primary_assembly.Pf3D7_01_v3.fa.gz
+
+log = snakemake.log_fmt_shell(stdout=False, stderr=True)
+
+suffix = ""
+if out_fmt == "gtf":
+    suffix = "gtf.gz"
+elif out_fmt == "gff3":
+    suffix = "gff3.gz"
+else:
+    raise ValueError(
+        "invalid format specified. Only 'gtf[.gz]' and 'gff3[.gz]' are currently supported."
+    )
+
+url = snakemake.params.get("url", "https://ftp.ebi.ac.uk/ensemblgenomes/pub/")
+url = f"{url}/release-{release}/{division}/{out_fmt}/{species}/{species.capitalize()}.{assembly}.{release}.{suffix}"
+ftp_url = url.replace("https://", "ftp://")
+
+try:
+    if out_gz:
+        shell("curl --fail -L {url} > {snakemake.output[0]} {log}")
+    else:
+        shell("(curl --fail -L {url} | gzip -d > {snakemake.output[0]}) {log}")
+except sp.CalledProcessError:
+    try:
+        if out_gz:
+            shell("curl --fail -L {ftp_url} > {snakemake.output[0]} {log}")
+        else:
+            shell("(curl --fail -L {ftp_url} | gzip -d > {snakemake.output[0]}) {log}")
+    except sp.CalledProcessError:
+        if snakemake.log:
+            sys.stderr = open(snakemake.log[0], "a")
+        print(
+            "Unable to download annotation data from Ensembl Genomes.\n"
+            "Did you check that this combination of species, division, assembly, and release is actually provided?",
+            file=sys.stderr,
+        )
+        exit(1)

bio/reference/ensembl-genomes-sequence/environment.linux-64.pin.txt

@@ -0,0 +1,26 @@
+# This file may be used to create an environment using:
+# $ conda create --name <env> --file <this file>
+# platform: linux-64
+# created-by: conda 25.3.1
+@EXPLICIT
+https://conda.anaconda.org/conda-forge/linux-64/_libgcc_mutex-0.1-conda_forge.tar.bz2#d7c89558ba9fa0495403155b64376d81
+https://conda.anaconda.org/conda-forge/noarch/ca-certificates-2025.11.12-hbd8a1cb_0.conda#f0991f0f84902f6b6009b4d2350a83aa
+https://conda.anaconda.org/conda-forge/linux-64/libgomp-15.2.0-he0feb66_16.conda#26c46f90d0e727e95c6c9498a33a09f3
+https://conda.anaconda.org/conda-forge/linux-64/_openmp_mutex-4.5-2_gnu.tar.bz2#73aaf86a425cc6e73fcf236a5a46396d
+https://conda.anaconda.org/conda-forge/linux-64/libgcc-15.2.0-he0feb66_16.conda#6d0363467e6ed84f11435eb309f2ff06
+https://conda.anaconda.org/conda-forge/linux-64/c-ares-1.34.6-hb03c661_0.conda#920bb03579f15389b9e512095ad995b7
+https://conda.anaconda.org/conda-forge/linux-64/keyutils-1.6.3-hb9d3cd8_0.conda#b38117a3c920364aff79f870c984b4a3
+https://conda.anaconda.org/conda-forge/linux-64/libgcc-ng-15.2.0-h69a702a_16.conda#5a68259fac2da8f2ee6f7bfe49c9eb8b
+https://conda.anaconda.org/conda-forge/linux-64/libstdcxx-15.2.0-h934c35e_16.conda#68f68355000ec3f1d6f26ea13e8f525f
+https://conda.anaconda.org/conda-forge/linux-64/libzlib-1.3.1-hb9d3cd8_2.conda#edb0dca6bc32e4f4789199455a1dbeb8
+https://conda.anaconda.org/conda-forge/linux-64/ncurses-6.5-h2d0b736_3.conda#47e340acb35de30501a76c7c799c41d7
+https://conda.anaconda.org/conda-forge/linux-64/openssl-3.6.0-h26f9b46_0.conda#9ee58d5c534af06558933af3c845a780
+https://conda.anaconda.org/conda-forge/linux-64/libedit-3.1.20250104-pl5321h7949ede_0.conda#c277e0a4d549b03ac1e9d6cbbe3d017b
+https://conda.anaconda.org/conda-forge/linux-64/libev-4.33-hd590300_2.conda#172bf1cd1ff8629f2b1179945ed45055
+https://conda.anaconda.org/conda-forge/linux-64/libssh2-1.11.1-hcf80075_0.conda#eecce068c7e4eddeb169591baac20ac4
+https://conda.anaconda.org/conda-forge/linux-64/libstdcxx-ng-15.2.0-hdf11a46_16.conda#1b3152694d236cf233b76b8c56bf0eae
+https://conda.anaconda.org/conda-forge/linux-64/zstd-1.5.7-hb78ec9c_6.conda#4a13eeac0b5c8e5b8ab496e6c4ddd829
+https://conda.anaconda.org/conda-forge/linux-64/krb5-1.21.3-h659f571_0.conda#3f43953b7d3fb3aaa1d0d0723d91e368
+https://conda.anaconda.org/conda-forge/linux-64/libnghttp2-1.67.0-had1ee68_0.conda#b499ce4b026493a13774bcf0f4c33849
+https://conda.anaconda.org/conda-forge/linux-64/libcurl-8.17.0-h4e3cde8_1.conda#117499f93e892ea1e57fdca16c2e8351
+https://conda.anaconda.org/conda-forge/linux-64/curl-8.17.0-h4e3cde8_1.conda#cb3648a22e7d5010fc9edae69bb2eab8

bio/reference/ensembl-genomes-sequence/environment.yaml

@@ -0,0 +1,5 @@
+channels:
+  - conda-forge
+  - nodefaults
+dependencies:
+  - curl

bio/reference/ensembl-genomes-sequence/meta.yaml

@@ -0,0 +1,14 @@
+name: ensembl-genomes-sequence
+description: |
+  Download sequences (e.g. genome) of non-vertebrate species from Ensembl
+  Genomes FTP servers, and store them in a single .fasta file. The release
+  numbers and folder organization differ from the Ensembl project, which
+  maintains data for vertebrate species. Based on wrapper ``ensembl-sequence``
+authors:
+  - Brandon Seah
+  - Johannes Köster
+url: https://ensemblgenomes.org/
+output:
+  - fasta file
+params:
+  - url: URL from where to download cache data (optional; by default is ``ftp://ftp.ebi.ac.uk/ensemblgenomes/pub/``)

bio/reference/ensembl-genomes-sequence/test/Snakefile

@@ -0,0 +1,81 @@
+rule get_genome:
+    output:
+        "refs/ensembl-genomes-genome.fasta",
+    params:
+        species="plasmodium_falciparum",
+        assembly="GCA000002765v3",
+        division="protists",
+        datatype="dna",
+        release="62",
+    log:
+        "logs/get_genome.log",
+    cache: "omit-software"  # save space and time with between workflow caching (see docs)
+    wrapper:
+        "master/bio/reference/ensembl-genomes-sequence"
+
+
+rule get_genome_gzipped:
+    output:
+        "refs/ensembl-genomes-genome.fa.gz",
+    params:
+        species="plasmodium_falciparum",
+        assembly="GCA000002765v3",
+        division="protists",
+        datatype="dna",
+        release="62",
+    log:
+        "logs/get_genome.log",
+    cache: "omit-software"  # save space and time with between workflow caching (see docs)
+    wrapper:
+        "master/bio/reference/ensembl-genomes-sequence"
+
+
+rule get_single_chromosome:
+    output:
+        "refs/ensembl-genomes-chrMt.fasta",
+    params:
+        species="schizosaccharomyces_pombe",
+        assembly="ASM294v2",
+        division="fungi",
+        datatype="dna",
+        release="62",
+        chromosome=["MT"],  # optional: restrict to one or multiple chromosomes, for multiple see below
+    log:
+        "logs/get_genome.log",
+    cache: "omit-software"  # save space and time with between workflow caching (see docs)
+    wrapper:
+        "master/bio/reference/ensembl-genomes-sequence"
+
+
+rule get_multiple_chromosome:
+    output:
+        "refs/ensembl-genomes-chr3_and_chrMt.fasta",
+    params:
+        species="schizosaccharomyces_pombe",
+        assembly="ASM294v2",
+        division="fungi",
+        datatype="dna",
+        release="62",
+        chromosome=["III", "MT"],
+    log:
+        "logs/get_genome.log",
+    cache: "omit-software"  # save space and time with between workflow caching (see docs)
+    wrapper:
+        "master/bio/reference/ensembl-genomes-sequence"
+
+
+rule get_multiple_chromosomes_gzipped:
+    output:
+        "refs/ensembl-genomes-chr3_and_chrMt.fasta.gz",
+    params:
+        species="schizosaccharomyces_pombe",
+        assembly="ASM294v2",
+        division="fungi",
+        datatype="dna",
+        release="62",
+        chromosome=["III", "MT"],
+    log:
+        "logs/get_genome.log",
+    cache: "omit-software"  # save space and time with between workflow caching (see docs)
+    wrapper:
+        "master/bio/reference/ensembl-genomes-sequence"