Merge branch 'master' into master

.github/scripts/wave_singularity.py

@@ -2,53 +2,86 @@
 # /// script
 # requires-python = ">=3.10"
 # dependencies = [
-#     "httpx",
+#     "requests",
+#     "rich",
 # ]
 # ///
 
 import logging
 
-import httpx
+import requests
+import rich_click as click
+from rich.logging import RichHandler
 
+# Replace the basic logger setup with rich logging
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(message)s",
+    handlers=[
+        RichHandler(
+            rich_tracebacks=True,
+            show_time=False,
+            markup=True,
+        )
+    ],
+)
 logger = logging.getLogger(__name__)
+click.rich_click.SHOW_ARGUMENTS = True
 
-image_url = "oras://community.wave.seqera.io/library/pybedtools_bedtools_htslib_pip_pypints:aa20de1f1b5ddb30"
-
-if image_url.startswith("oras://"):
-    image_url = image_url.replace("oras://", "")
-
-wave_api_url = "https://wave.seqera.io"
-url = f"{wave_api_url}/v1alpha1/inspect"
-
-# if platform_pat:
-#     data["toweraccesstoken"] = platform_pat
-# else:
-# TODO
-logger.warning("'platform_pat' not set, no auth to wave back end")
-
-try:
-    logger.info(f"calling image inspect at {url} for image url {image_url}")
-    response = httpx.post(
-        url=url,
-        json={"containerImage": image_url},
-        headers={"content-type": "application/json"},
-    )
-
-    data = response.json()
-    logger.debug(data)
-    layers = data.get("container", {}).get("manifest", {}).get("layers", [])
-    is_singularity = len(layers) == 1 and layers[0].get("mediaType", "").endswith(".sif")
-    if not is_singularity:
-        print(layers)
-        raise ValueError("not a singularity image")
-    if "digest" not in layers[0]:
-        print(layers)
-        raise ValueError("no 'digest' in first layer found")
-
-    digest = layers[0]["digest"].replace("sha256:", "")
-    container_url = f"https://community-cr-prod.seqera.io/docker/registry/v2/blobs/sha256/{digest[:2]}/{digest}/data"
-    print(container_url)
-
-except httpx.RequestError as exc:
-    print(f"An error occurred while requesting {exc.request.url!r}.")
-    print("No singularity image for you")
+
+@click.command()
+@click.option(
+    "--platform-pat",
+    envvar="SEQERA_ACCESS_TOKEN",
+    show_envvar=True,
+    help="Platform authentication token",
+)
+@click.argument("image_name")
+def main(image_name, platform_pat):
+    """Script to return a HTTPS Singularity image URL from Seqera Containers."""
+
+    if image_name.startswith("oras://"):
+        image_name = image_name.replace("oras://", "")
+
+    wave_api_url_base = "https://wave.seqera.io"
+    wave_api_url = f"{wave_api_url_base}/v1alpha1/inspect"
+    container_url_base = "https://community-cr-prod.seqera.io/docker/registry/v2/blobs/sha256"
+
+    request_data = {"containerImage": image_name}
+    if platform_pat:
+        request_data["toweraccesstoken"] = platform_pat
+    else:
+        logger.debug("'--platform-pat' / '$TOWER_ACCESS_TOKEN' not set, no auth to wave back end")
+
+    try:
+        logger.debug(f"Calling image inspect at {wave_api_url} for image url '{image_name}'")
+        response = requests.post(
+            url=wave_api_url,
+            json=request_data,
+            headers={"content-type": "application/json"},
+        )
+
+        data = response.json()
+        logger.debug(data)
+        layers = data.get("container", {}).get("manifest", {}).get("layers", [])
+        is_singularity = len(layers) == 1 and layers[0].get("mediaType", "").endswith(".sif")
+        logger.debug(layers)
+        if not is_singularity:
+            raise ValueError("Not a singularity image")
+        if "digest" not in layers[0]:
+            raise ValueError("no 'digest' in first layer found")
+
+        digest = layers[0]["digest"].replace("sha256:", "")
+        container_url = f"{container_url_base}/{digest[:2]}/{digest}/data"
+        print(container_url)
+
+    except requests.RequestException as exc:
+        raise ValueError(f"An error occurred while requesting {wave_api_url}\n {exc}")
+
+
+if __name__ == "__main__":
+    try:
+        main()
+    except ValueError as exc:
+        logger.error(f"[red]{exc}[/red]")
+        exit(1)

modules/nf-core/metamdbg/asm/environment.yml

@@ -0,0 +1,7 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  - "bioconda::metamdbg=1.0"

modules/nf-core/metamdbg/asm/main.nf

@@ -0,0 +1,62 @@
+process METAMDBG_ASM {
+    tag "${meta.id}"
+    label 'process_medium'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/metamdbg:1.0--hdcf5f25_1':
+        'biocontainers/metamdbg:1.0--hdcf5f25_1' }"
+
+    input:
+    tuple val(meta), path(reads)
+    val(input_type)
+
+    output:
+    tuple val(meta), path("*.contigs.fasta.gz"), emit: contigs
+    tuple val(meta), path("*.metaMDBG.log")    , emit: log
+    path "versions.yml"                        , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    switch(input_type) {
+        case "hifi": input = "--in-hifi ${reads}"; break
+        case "ont" : input = "--in-ont ${reads}" ; break
+        default:
+            error("ERROR: input_type must be one of either 'hifi' or 'ont'.")
+            break
+    }
+    """
+    metaMDBG asm \\
+        --threads ${task.cpus} \\
+        --out-dir . \\
+        ${args} \\
+        ${input}
+
+    rm -r tmp/
+
+    mv contigs.fasta.gz ${prefix}.contigs.fasta.gz
+    mv metaMDBG.log ${prefix}.metaMDBG.log
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        metamdbg: \$(metaMDBG | grep "Version" | sed 's/ Version: //')
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch ${prefix}.metaMDBG.log
+    touch ${prefix}.contigs.fasta.gz
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        metamdbg: \$(metaMDBG | grep "Version" | sed 's/ Version: //')
+    END_VERSIONS
+    """
+}

modules/nf-core/metamdbg/asm/meta.yml

@@ -0,0 +1,67 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/meta-schema.json
+name: "METAMDBG_ASM"
+description: Metagenome assembler for long-read sequences (HiFi and ONT).
+keywords:
+  - assembly
+  - long reads
+  - metagenome
+  - metagenome assembler
+tools:
+  - "metamdbg":
+      description: "MetaMDBG: a lightweight assembler for long and accurate metagenomics
+        reads."
+      homepage: "https://github.com/GaetanBenoitDev/metaMDBG"
+      documentation: "https://github.com/GaetanBenoitDev/metaMDBG"
+      tool_dev_url: "https://github.com/GaetanBenoitDev/metaMDBG"
+      doi: "10.1038/s41587-023-01983-6"
+      licence: ["MIT"]
+      identifier: ""
+
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1', single_end:false ]`
+
+    - reads:
+        type: file
+        description: Long read sequence data from ONT or HiFi in fasta format (can be
+          gzipped)
+        pattern: "*.{fa,fasta,fa.gz,fasta.gz}"
+
+  - - input_type:
+        type: string
+        description: Sequencing technology for reads - either "hifi" for PacBio HiFi
+          reads or "ont" for Oxford Nanopore reads.
+
+output:
+  - contigs:
+      - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1', single_end:false ]`
+      - "*.contigs.fasta.gz":
+          type: file
+          description: |
+            Gzipped fasta file containing the assembled contigs from the input
+            reads.
+  - log:
+      - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1', single_end:false ]`
+      - "*.metaMDBG.log":
+          type: file
+          description: Log file describing the metaMDBG run.
+  - versions:
+      - versions.yml:
+          type: file
+          description: File containing software versions
+          pattern: "versions.yml"
+authors:
+  - "@prototaxites"
+maintainers:
+  - "@prototaxites"

modules/nf-core/metamdbg/asm/tests/main.nf.test

@@ -0,0 +1,116 @@
+nextflow_process {
+
+    name "Test Process METAMDBG_ASM"
+    script "../main.nf"
+    process "METAMDBG_ASM"
+
+    tag "modules"
+    tag "modules_nfcore"
+    tag "metamdbg"
+    tag "metamdbg/asm"
+
+    test("metamdbg_asm - ont") {
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/prokaryotes/bacteroides_fragilis/nanopore/fastq/test.fastq.gz', checkIfExists: true),
+                ]
+                input[1] = "ont"
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                // output is stochastic - contig names differ per run
+                // log file contains nextflow work dir paths
+                { assert snapshot(
+                    file(process.out.contigs[0][1]).name,
+                    file(process.out.log[0][1]).name,
+                    process.out.versions
+                ).match() },
+                { assert path(process.out.log[0][1]).readLines().last().contains("Done!") }
+            )
+        }
+    }
+    test("metamdbg_asm - hifi") {
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/fastq/test_hifi.fastq.gz', checkIfExists: true),
+                ]
+                input[1] = "hifi"
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                // output is stochastic - contig names differ per run
+                // log file contains nextflow work dir paths
+                { assert snapshot(
+                    file(process.out.contigs[0][1]).name,
+                    file(process.out.log[0][1]).name,
+                    process.out.versions
+                ).match() },
+                { assert path(process.out.log[0][1]).readLines().last().contains("Done!") }
+            )
+        }
+    }
+
+    test("metamdbg_asm - wrong format") {
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/fastq/test_hifi.fastq.gz', checkIfExists: true),
+                ]
+                input[1] = "wrong"
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.failed },
+                { assert process.errorReport.contains("ERROR: input_type must be one of either 'hifi' or 'ont'.") }
+            )
+        }
+
+    }
+
+    test("metamdbg_asm - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/pacbio/fastq/test_hifi.fastq.gz', checkIfExists: true),
+                ]
+                input[1] = "hifi"
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out.versions).match("stub_versions") }
+            )
+        }
+
+    }
+}