Merge pull request #35 from Ferlab-Ste-Justine/feat/allow-older-gatk-…

…versions-for-test-dataset

feat: Allow older GATK versions for test dataset

.github/workflows/ci.yml

@@ -42,4 +42,8 @@ jobs:
 
         # Using the stub mode until our test data setup is ready
         run: |
-          nextflow run ${GITHUB_WORKSPACE} -stub -profile test,docker --input assets/TestSampleSheet.csv
+          nextflow run ${GITHUB_WORKSPACE} -stub-run -profile test,docker --input "${GITHUB_WORKSPACE}/assets/TestSampleSheet.csv"
+
+      - name: Display nextflow.log on error
+        if: failure()
+        run: cat .nextflow.log

CHANGELOG.md

@@ -5,6 +5,14 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## v2.1.0dev - [date]
 
+### `Added`
+- [#35](https://github.com/Ferlab-Ste-Justine/Post-processing-Pipeline/pull/35) Added allow_old_gatk_data parameter (set to false by default).
+- [#35](https://github.com/Ferlab-Ste-Justine/Post-processing-Pipeline/pull/35) Added missing stub block in process writemeta for compatibility with latest nextflow version.
+- [#35](https://github.com/Ferlab-Ste-Justine/Post-processing-Pipeline/pull/35) Improve github ci workflow to display nextflow log file content on error
+
+### `Fixed`
+- [#35](https://github.com/Ferlab-Ste-Justine/Post-processing-Pipeline/pull/35) Fix incorrect assumption about assets folder location in github ci workflow
+
 
 ## v2.0.0dev
 

conf/modules.config

@@ -36,6 +36,7 @@ process {
 
     withName: GATK4_GENOTYPEGVCFS {
         container = 'broadinstitute/gatk:4.5.0.0'
+        ext.args = params.allow_old_gatk_data ? "--allow-old-rms-mapping-quality-annotation-data" : ""
         ext.prefix = {meta.id + ".genotyped"}
     }
 

conf/test.config

@@ -12,7 +12,7 @@
     archive (.tar.gz) and extract it at the root of this  repository (tar -xzf <archive file>), you should 
     be able to run the pipeline with the following command:
 
-        nextflow -c data-test/test.config  run  Ferlab-Ste-Justine/Post-processing-Pipeline -r main -profile test,docker
+        nextflow run  Ferlab-Ste-Justine/Post-processing-Pipeline -r main -profile test,docker
     
 ----------------------------------------------------------------------------------------
 */
@@ -60,4 +60,7 @@ params {
     exomiser_remm_filename = "ReMM.v0.3.1.post1.hg38.tsv.gz"
     exomiser_analysis_wes = "${projectDir}/assets/exomiser/test_exomiser_analysis.yml"
     exomiser_analysis_wgs = "${projectDir}/assets/exomiser/test_exomiser_analysis.yml"
-}
+
+    // To be able to run on our public test dataset, which is aligned with an older version of GATK4
+    allow_old_gatk_data = true
+}

nextflow.config

@@ -33,7 +33,6 @@ params {
     exomiser_analysis_wes = "${projectDir}/assets/exomiser/default_exomiser_WES_analysis.yml"
     exomiser_analysis_wgs = "${projectDir}/assets/exomiser/default_exomiser_WGS_analysis.yml"
 
-
 	//Process-specific parameters
     TSfilterSNP = '99'
     TSfilterINDEL = '99'
@@ -45,6 +44,9 @@ params {
                    [name: 'MQ40', expression: 'MQ < 40.0'],
                    [name: 'MQRankSum-12.5', expression: 'MQRankSum < -12.5'],
                    [name: 'ReadPosRankSum-8', expression: 'ReadPosRankSum < -8.0']]
+
+
+    allow_old_gatk_data = false
 
 	//Resources optionsreferenceGenome
 	//defaults expecting to be overwritten

nextflow_schema.json

@@ -251,6 +251,11 @@
           "description": "Parameters for Hard-Filtering",
           "help_text": "Parameters for Hard-Filtering. Must be an array containing each desired filter. Each filter must be formatted with the desired name and expression, for example\\n[[name: 'QD1', expression: 'QD < 1.0'],[name: 'QD2', expression: 'QD < 2.0]]"
         },
+        "allow_old_gatk_data": {
+          "type": "boolean",
+          "description": "Allow the use of old GATK data in GATK_GENOTYPEGVCFS process by adding --allow-old-rms-mapping-quality-annotation-data to ext.args. Note: this won't work if ext.args is overridden.",
+          "help_text": "Allow the use of old GATK data in GATK_GENOTYPEGVCFS process by adding --allow-old-rms-mapping-quality-annotation-data to ext.args. Note that this won't work if ext.args is overridden. This parameter is not recommended for production use."
+        },
         "tools": {
           "type": "string",
           "pattern": "^(vep|exomiser)?(,(vep|exomiser))*$",
@@ -374,7 +379,10 @@
             "required": ["exomiser_cadd_version", "tools"]
           },
           "then": {
-            "required": ["exomiser_cadd_indel_filename", "exomiser_cadd_snv_filename"]
+            "required": [
+              "exomiser_cadd_indel_filename",
+              "exomiser_cadd_snv_filename"
+            ]
           }
         },
         {

subworkflows/local/utils_nfcore_postprocessing_pipeline/main.nf

@@ -192,6 +192,9 @@ def getPipelineInfoFolder(outdir) {
 // Check and validate pipeline parameters
 //
 def validateInputParameters() {
+    if (params.allow_old_gatk_data) {
+        log.warn "The 'allow_old_gatk_data' parameter is set to true, allowing the pipeline to run with older GATK data in GATK4_GENOTYPEGVCFS. Not recommended for production."
+    }
     genomeExistsError()
 }
 

workflows/postprocessing.nf

@@ -104,6 +104,11 @@ process writemeta{
     Revision : ${workflow.revision}
     CommitId : ${workflow.commitId}
     """
+
+    stub:
+    """
+    touch metadata.txt
+    """
 }