Module to register study for assembly uploads

.gitignore

@@ -7,5 +7,6 @@ testing/
 testing*
 *.pyc
 null/
+.nf-test*
 .idea/
 test_data

modules/local/registerstudy/environment.yml

@@ -0,0 +1,10 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/environment-schema.json
+channels:
+  - conda-forge
+  - bioconda
+dependencies:
+  # TODO nf-core: List required Conda package(s).
+  #               Software MUST be pinned to channel (i.e. "bioconda"), version (i.e. "1.10").
+  #               For Conda, the build (i.e. "h9402c20_2") must be EXCLUDED to support installation on different operating systems.
+  - "bioconda::assembly_uploader=1.3.2"

modules/local/registerstudy/main.nf

@@ -0,0 +1,55 @@
+process REGISTERSTUDY {
+    tag "$meta.id"
+    label 'process_single'
+
+    conda "${moduleDir}/environment.yml"
+    container "community.wave.seqera.io/library/pip_assembly-uploader:7e9461afbdd7a521"
+
+    input:
+    tuple val(meta), val(study), val(center), val(library)
+
+
+    output:
+    tuple val(meta), env(STUDY_ID), emit: study_accession
+    path "versions.yml"           , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def args2 = task.ext.args2 ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    study_xmls \\
+        $args \\
+        --study ${study} \\
+        --library ${library} \\
+        --center ${center} \\
+
+
+    submit_study \\
+        $args2 \\
+        --directory ${study}_upload \\
+        --study ${study} 2>&1 | tee report.log
+
+    STUDY_ID=\$(grep 'A new study accession has been created' report.log | grep -oE '(PRJ|ERP)[[:alnum:]_]+[[:digit:]]+')
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        python: \$(python --version)
+    END_VERSIONS
+    """
+
+    stub:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch ${prefix}.report
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        python: \$(python --version)
+    END_VERSIONS
+    """
+}

modules/local/registerstudy/meta.yml

@@ -0,0 +1,70 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/meta-schema.json
+name: "registerstudy"
+description: This module registers a study and project and generates accessions that will be used for metagenomic assembly uploads in ENA. The study generated will reference reads from an already public project.
+keywords:
+  - assembly
+  - register
+  - study
+tools:
+  - "registerstudy":
+      description: "Nextflow module to register study/project to upload primary metagenome and metatranscriptome
+        assemblies to ENA on a per-study basis. The scripts generate xmls to register a new study and create manifests
+        necessary for submission of assemblies using webin-cli."
+      homepage: "https://github.com/EBI-Metagenomics/assembly_uploader"
+      documentation: "https://github.com/EBI-Metagenomics/assembly_uploader"
+      tool_dev_url: "None"
+      doi: ""
+      licence: ["Apache-2.0"]
+      identifier: null
+
+input:
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1' ]`
+    - study:
+        type: value
+        description: |
+          Study accession with raw reads public in ENA.
+          Example: "PRJNA312520"
+
+    - center:
+        type: value
+        description: |
+          Name of the sequencing or submitting center.
+          Example: "Wellcome Sanger Institute"
+
+  - library:
+      type: value
+      description: |
+        Library information associated with the study.
+        Example: "metagenome"
+      enum:
+        - metagenome
+        - metatranscriptome
+
+output:
+  study_accession:
+    - - meta:
+          type: map
+          description: |
+            Groovy Map containing sample information
+            e.g. `[ id:'sample1' ]`
+      - study:
+          type: value
+          description: |
+            Study accession registered in ENA.
+            Example: "PRJEB312520"
+  versions:
+    - "versions.yml":
+        type: file
+        description: File containing software versions
+        pattern: "versions.yml"
+        ontologies:
+          - edam: "http://edamontology.org/format_3750" # YAML
+
+authors:
+  - "@alisha246"
+maintainers:
+  - "@alisha246"

modules/local/registerstudy/nextflow.config

@@ -0,0 +1,5 @@
+process {
+    withName: REGISTERSTUDY {
+        ext.args2 = '--test'
+    }
+}

modules/local/registerstudy/tests/main.nf.test

@@ -0,0 +1,63 @@
+// TODO nf-core: Once you have added the required tests, please run the following command to build this file:
+// nf-core modules test registerstudy
+nextflow_process {
+    name "Test Process REGISTERSTUDY"
+    script "../main.nf"
+    config "../nextflow.config"
+    process "REGISTERSTUDY"
+
+
+    tag "modules"
+    tag "registerstudy"
+/*
+    test("registerstudy - should register a study on ENA test server") {
+
+        when {
+            process {
+               """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    "PRJNA318468",
+                    "EMG",
+                    "metagenome"
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                //TODO improve assertions
+            )
+        }
+    }
+*/
+
+    test("registerstudy - stub") {
+
+        options "-stub"
+
+        when {
+            process {
+                """
+                input[0] = [
+                    [ id:'test', single_end:false ], // meta map
+                    "PRJNA318468",
+                    "EMG",
+                    "metagenome"
+                ]
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(process.out).match() }
+                //TODO improve assertions
+            )
+        }
+
+    }
+}

modules/local/registerstudy/tests/main.nf.test.snap

@@ -0,0 +1,72 @@
+{
+    "registerstudy - report - stub": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        ""
+                    ]
+                ],
+                "1": [
+                    "versions.yml:md5,ea872d341a2054fde3b2c8f06bbf8177"
+                ],
+                "study_accession": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        ""
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,ea872d341a2054fde3b2c8f06bbf8177"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.3",
+            "nextflow": "25.10.0"
+        },
+        "timestamp": "2025-10-28T16:35:02.331026"
+    },
+    "registerstudy - stub": {
+        "content": [
+            {
+                "0": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        ""
+                    ]
+                ],
+                "1": [
+                    "versions.yml:md5,049c28351d1fd639673904f17b95c213"
+                ],
+                "study_accession": [
+                    [
+                        {
+                            "id": "test",
+                            "single_end": false
+                        },
+                        ""
+                    ]
+                ],
+                "versions": [
+                    "versions.yml:md5,049c28351d1fd639673904f17b95c213"
+                ]
+            }
+        ],
+        "meta": {
+            "nf-test": "0.9.3",
+            "nextflow": "25.10.0"
+        },
+        "timestamp": "2025-10-29T10:59:32.499098"
+    }
+}