Retrieving module information of input and output and EDAM ontology from bio.tools

.github/workflows/pytest.yml

@@ -75,8 +75,8 @@ jobs:
     name: Run ${{matrix.test}} with Python ${{ needs.setup.outputs.python-version }} on ${{ needs.setup.outputs.runner }}
     needs: [setup, list_tests]
     if: ${{ needs.setup.outputs.run-tests }}
-    # run on self-hosted runners for test_components.py (because of the gitlab branch), based on the input if it is dispatched manually, on github if it is a rerun or on self-hosted by default
-    runs-on: ${{ matrix.test == 'test_components.py' && 'self-hosted' || (github.event.inputs.runners || github.run_number > 1 && 'ubuntu-latest' || 'self-hosted') }}
+    # run on self-hosted runners for test_components_generate_snapshot.py (because of the gitlab branch), based on the input if it is dispatched manually, on github if it is a rerun or on self-hosted by default
+    runs-on: ${{ matrix.test == 'components/test_components_generate_snapshot.py' && 'self-hosted' || (github.event.inputs.runners || github.run_number > 1 && 'ubuntu-latest' || 'self-hosted') }}
     strategy:
       matrix: ${{ fromJson(needs.list_tests.outputs.tests) }}
       fail-fast: false # run all tests even if one fails

nf_core/components/components_utils.py

@@ -1,7 +1,7 @@
 import logging
 import re
 from pathlib import Path
-from typing import TYPE_CHECKING, List, Optional, Tuple, Union
+from typing import TYPE_CHECKING, Dict, List, Optional, Tuple, Union
 
 import questionary
 import requests
@@ -165,9 +165,9 @@ def get_components_to_install(subworkflow_dir: Union[str, Path]) -> Tuple[List[s
     return modules, subworkflows
 
 
-def get_biotools_id(tool_name) -> str:
+def get_biotools_response(tool_name: str) -> Optional[Dict]:
     """
-    Try to find a bio.tools ID for 'tool'
+    Try to get bio.tools information for 'tool'
     """
     url = f"https://bio.tools/api/t/?q={tool_name}&format=json"
     try:
@@ -176,16 +176,74 @@ def get_biotools_id(tool_name) -> str:
         response.raise_for_status()  # Raise an error for bad status codes
         # Parse the JSON response
         data = response.json()
+        log.info(f"Found bio.tools information for '{tool_name}'")
+        return data
 
-        # Iterate through the tools in the response to find the tool name
-        for tool in data["list"]:
-            if tool["name"].lower() == tool_name:
-                return tool["biotoolsCURIE"]
+    except requests.exceptions.RequestException as e:
+        log.warning(f"Could not find bio.tools information for '{tool_name}': {e}")
+        return None
 
-        # If the tool name was not found in the response
-        log.warning(f"Could not find a bio.tools ID for '{tool_name}'")
-        return ""
 
-    except requests.exceptions.RequestException as e:
-        log.warning(f"Could not find a bio.tools ID for '{tool_name}': {e}")
-        return ""
+def get_biotools_id(data: dict, tool_name: str) -> str:
+    """
+    Try to find a bio.tools ID for 'tool'
+    """
+    # Iterate through the tools in the response to find the tool name
+    for tool in data["list"]:
+        if tool["name"].lower() == tool_name:
+            log.info(f"Found bio.tools ID: '{tool['biotoolsCURIE']}'")
+            return tool["biotoolsCURIE"]
+
+    # If the tool name was not found in the response
+    log.warning(f"Could not find a bio.tools ID for '{tool_name}'")
+    return ""
+
+
+DictWithStrAndTuple = Dict[str, Tuple[List[str], List[str], List[str]]]
+
+
+def get_channel_info_from_biotools(
+    data: dict, tool_name: str
+) -> Optional[Tuple[DictWithStrAndTuple, DictWithStrAndTuple]]:
+    """
+    Try to find input and output channels and the respective EDAM ontology terms
+
+    Args:
+        data (dict): The bio.tools API response
+        tool_name (str): The name of the tool
+    """
+    inputs = {}
+    outputs = {}
+
+    def _iterate_input_output(type) -> DictWithStrAndTuple:
+        type_info = {}
+        if type in funct:
+            for element in funct[type]:
+                if "data" in element:
+                    element_name = "_".join(element["data"]["term"].lower().split(" "))
+                    uris = [element["data"]["uri"]]
+                    terms = [element["data"]["term"]]
+                    patterns = []
+                if "format" in element:
+                    for format in element["format"]:
+                        # Append the EDAM URI
+                        uris.append(format["uri"])
+                        # Append the EDAM term, getting the first word in case of complicated strings. i.e. "FASTA format"
+                        patterns.append(format["term"].lower().split(" ")[0])
+                        terms.append(format["term"])
+                    type_info[element_name] = (uris, terms, patterns)
+        return type_info
+
+    # Iterate through the tools in the response to find the tool name
+    for tool in data["list"]:
+        if tool["name"].lower() == tool_name:
+            if "function" in tool:
+                # Parse all tool functions
+                for funct in tool["function"]:
+                    inputs.update(_iterate_input_output("input"))
+                    outputs.update(_iterate_input_output("output"))
+            return inputs, outputs
+
+    # If the tool name was not found in the response
+    log.warning(f"Could not find an EDAM ontology term for '{tool_name}'")
+    return None

nf_core/components/create.py

@@ -21,7 +21,7 @@
 import nf_core
 import nf_core.utils
 from nf_core.components.components_command import ComponentCommand
-from nf_core.components.components_utils import get_biotools_id
+from nf_core.components.components_utils import get_biotools_id, get_biotools_response, get_channel_info_from_biotools
 from nf_core.pipelines.lint_utils import run_prettier_on_file
 
 log = logging.getLogger(__name__)
@@ -151,8 +151,15 @@ def create(self) -> bool:
             if self.component_type == "modules":
                 # Try to find a bioconda package for 'component'
                 self._get_bioconda_tool()
+                name = self.tool_conda_name if self.tool_conda_name else self.component
                 # Try to find a biotools entry for 'component'
-                self.tool_identifier = get_biotools_id(self.component)
+                biotools_data = get_biotools_response(name)
+                if biotools_data:
+                    self.tool_identifier = get_biotools_id(biotools_data, name)
+                    # Obtain EDAM ontologies for inputs and outputs
+                    channel_info = get_channel_info_from_biotools(biotools_data, name)
+                    if channel_info:
+                        self.inputs, self.outputs = channel_info
 
             # Prompt for GitHub username
             self._get_username()
@@ -176,6 +183,8 @@ def create(self) -> bool:
 
         new_files = [str(path) for path in self.file_paths.values()]
 
+        run_prettier_on_file(new_files)
+
         log.info("Created following files:\n  " + "\n  ".join(new_files))
         return True
 

nf_core/module-template/environment.yml

@@ -4,4 +4,7 @@ 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 if bioconda else 'YOUR-TOOL-HERE' }}"

nf_core/module-template/main.nf

@@ -22,9 +22,6 @@ process {{ component_name_underscore|upper }} {
     label '{{ process_label }}'
 
     {% if not_empty_template -%}
-    // 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.
     // TODO nf-core: See section in main README for further information regarding finding and adding container addresses to the section below.
     {% endif -%}
     conda "${moduleDir}/environment.yml"
@@ -33,6 +30,12 @@ process {{ component_name_underscore|upper }} {
         '{{ docker_container if docker_container else 'biocontainers/YOUR-TOOL-HERE' }}' }"
 
     input:
+    {%- if inputs %}
+    // TODO nf-core: Update the information obtained from bio.tools and make sure that it is correct
+    {%- for input_name, ontologies in inputs.items() %}
+    {{ 'tuple val(meta), path(' + input_name + ')' if has_meta else 'path ' + input_name }}
+    {%- endfor %}
+    {%- else -%}
     {% if not_empty_template -%}
     // TODO nf-core: Where applicable all sample-specific information e.g. "id", "single_end", "read_group"
     //               MUST be provided as an input via a Groovy Map called "meta".
@@ -44,16 +47,22 @@ process {{ component_name_underscore|upper }} {
     {%- else -%}
     {{ 'tuple val(meta), path(input)' if has_meta else 'path input' }}
     {%- endif %}
+    {%- endif %}
 
     output:
+    {%- if outputs %}
+    // TODO nf-core: Update the information obtained from bio.tools and make sure that it is correct
+    {%- for output_name, ontologies in outputs.items() %}
+    {{ 'tuple val(meta), path("*.{' + ontologies[2]|join(',') + '}")' if has_meta else 'path ' + output_name }}, emit: {{ output_name }}
+    {%- endfor %}
+    {%- else %}
     {% if not_empty_template -%}
     // TODO nf-core: Named file extensions MUST be emitted for ALL output channels
     {{ 'tuple val(meta), path("*.bam")' if has_meta else 'path "*.bam"' }}, emit: bam
+    // TODO nf-core: List additional required output channels/values here
     {%- else -%}
     {{ 'tuple val(meta), path("*")' if has_meta else 'path "*"' }}, emit: output
     {%- endif %}
-    {% if not_empty_template -%}
-    // TODO nf-core: List additional required output channels/values here
     {%- endif %}
     path "versions.yml"           , emit: versions
 
@@ -78,20 +87,33 @@ process {{ component_name_underscore|upper }} {
     {%- endif %}
     """
     {% if not_empty_template -%}
-    samtools \\
-        sort \\
+    {{ component }} \\
         $args \\
         -@ $task.cpus \\
         {%- if has_meta %}
+        {%- if inputs %}
+        {%- for input_name, ontologies in inputs.items() %}
+        {%- set extensions = ontologies[2] %}
+        {%- for ext in extensions %}
+        -o ${prefix}.{{ ext }} \\
+        {%- endfor %}
+        {%- endfor %}
+        {%- else %}
         -o ${prefix}.bam \\
-        -T $prefix \\
         {%- endif %}
+        {%- endif %}
+        {%- if inputs %}
+        {%- for input_name, ontologies in inputs.items() %}
+        ${{ input_name }} \\
+        {%- endfor %}
+        {%- else %}
         $bam
+        {%- endif %}
     {%- endif %}
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":
-        {{ component }}: \$(samtools --version |& sed '1!d ; s/samtools //')
+        {{ component }}: \$({{ component }} --version)
     END_VERSIONS
     """
 
@@ -108,12 +130,21 @@ process {{ component_name_underscore|upper }} {
     {%- endif %}
     """
     {% if not_empty_template -%}
+    {%- if inputs %}
+    {%- for input_name, ontologies in inputs.items() %}
+    {%- set extensions = ontologies[2] %}
+    {%- for ext in extensions %}
+    touch ${prefix}.{{ ext }}
+    {%- endfor %}
+    {%- endfor %}
+    {%- else %}
     touch ${prefix}.bam
     {%- endif %}
+    {%- endif %}
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":
-        {{ component }}: \$(samtools --version |& sed '1!d ; s/samtools //')
+        {{ component }}: \$({{ component }} --version)
     END_VERSIONS
     """
 }

nf_core/module-template/meta.yml

@@ -26,12 +26,32 @@ tools:
 ## TODO nf-core: Add a description of all of the variables used as input
 {% endif -%}
 input:
+  {% if inputs -%}
+  {% for input_name, ontologies in inputs.items() -%}
+  {% if has_meta %}
+  - - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1' ]`
+  {% endif %}
+    - {{ input_name }}:
+        # TODO nf-core: Update the information obtained from bio.tools and make sure that it is correct
+        type: file
+        description: {{ input_name }} file
+        pattern: {{ "\"*.{" + ontologies[2]|join(",") + "}\"" }}
+        ontologies:
+          {% for ontology in ontologies[0] -%}
+          - edam: "{{ ontology }}" # {{ ontologies[1][loop.index0] }}
+          {% endfor -%}
+  {% endfor -%}
+  {% else -%}
   #{% if has_meta %} Only when we have meta
   - - meta:
         type: map
         description: |
           Groovy Map containing sample information
-          e.g. `[ id:'sample1', single_end:false ]`
+          e.g. `[ id:'sample1' ]`
   {% endif %}
   {% if not_empty_template -%}
   ## TODO nf-core: Delete / customise this example input
@@ -42,24 +62,46 @@ input:
         pattern: {{ '"*.{bam,cram,sam}"' if not_empty_template else "" }}
         ontologies:
           {% if not_empty_template -%}
-          - edam: "http://edamontology.org/format_25722"
-          - edam: "http://edamontology.org/format_2573"
-          - edam: "http://edamontology.org/format_3462"
-          {% else %}
+          - edam: "http://edamontology.org/format_25722" # BAM
+          - edam: "http://edamontology.org/format_2573" # CRAM
+          - edam: "http://edamontology.org/format_3462" # SAM
+          {% else -%}
           - edam: ""
           {%- endif %}
+  {%- endif %}
 
 {% if not_empty_template -%}
 ## TODO nf-core: Add a description of all of the variables used as output
 {% endif -%}
 output:
+  {% if outputs -%}
+  {% for output_name, ontologies in outputs.items() -%}
+  - {{ output_name }}:
+    {% if has_meta -%}
+    - meta:
+        type: map
+        description: |
+          Groovy Map containing sample information
+          e.g. `[ id:'sample1' ]`
+    {%- endif %}
+    - {{ "\"*.{" + ontologies[2]|join(",") + "}\"" }}:
+        # TODO nf-core: Update the information obtained from bio.tools and make sure that it is correct
+        type: file
+        description: {{ output_name }} file
+        pattern: {{ "\"*.{" + ontologies[2]|join(",") + "}\"" }}
+        ontologies:
+          {%- for ontology in ontologies[0] %}
+          - edam: "{{ ontology }}" # {{ ontologies[1][loop.index0] }}
+          {%- endfor %}
+  {% endfor -%}
+  {% else -%}
   - {{ 'bam:' if not_empty_template else "output:" }}
   #{% if has_meta -%} Only when we have meta
     - meta:
         type: map
         description: |
           Groovy Map containing sample information
-          e.g. `[ id:'sample1', single_end:false ]`
+          e.g. `[ id:'sample1' ]`
   {%- endif %}
   {% if not_empty_template -%}
     ## TODO nf-core: Delete / customise this example output
@@ -70,12 +112,13 @@ output:
         pattern: {{ '"*.{bam,cram,sam}"' if not_empty_template else "" }}
         ontologies:
           {% if not_empty_template -%}
-          - edam: "http://edamontology.org/format_25722"
-          - edam: "http://edamontology.org/format_2573"
-          - edam: "http://edamontology.org/format_3462"
+          - edam: "http://edamontology.org/format_25722" # BAM
+          - edam: "http://edamontology.org/format_2573" # CRAM
+          - edam: "http://edamontology.org/format_3462" # SAM
           {% else -%}
           - edam: ""
           {%- endif %}
+  {%- endif %}
   - versions:
     - "versions.yml":
         type: file