Implement panel resource creation workflow

conf/modules.config

@@ -274,4 +274,28 @@ process {
         ]
     }
 
+    withName: 'COBALT_PANEL_NORMALISATION' {
+        publishDir = [
+            path: { "${params.outdir}" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : "panel_resources/${filename}" },
+        ]
+    }
+
+    withName: 'PAVE_PON_PANEL_CREATION' {
+        publishDir = [
+            path: { "${params.outdir}" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : "panel_resources/${filename}" },
+        ]
+    }
+
+    withName: 'ISOFOX_PANEL_NORMALISATION' {
+        publishDir = [
+            path: { "${params.outdir}" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : "panel_resources/${filename}" },
+        ]
+    }
+
 }

conf/panel_resource_creation_parameters.config

@@ -0,0 +1,5 @@
+process {
+    withName: '^.*:AMBER' {
+        ext.args = '-tumor_min_depth 2'
+    }
+}

lib/Constants.groovy

@@ -25,6 +25,7 @@ class Constants {
 
 
     static enum RunMode {
+        PANEL_RESOURCE_CREATION,
         TARGETED,
         WGTS,
     }

lib/Utils.groovy

@@ -408,6 +408,15 @@ class Utils {
             Nextflow.exit(1)
         }
 
+        // Require --isofox_gene_ids argument to be provided in PANEL_RESOURCE_CREATION when RNA inputs are present
+        if (run_config.mode === Constants.RunMode.PANEL_RESOURCE_CREATION && run_config.has_rna && !params.isofox_gene_ids) {
+            log.error "\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n" +
+                "  Running the panel resource creation workflow with RNA requires that the\n" +
+                "  --isofox_gene_ids argument is set with an appropriate input file.\n" +
+                "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~"
+            Nextflow.exit(1)
+        }
+
     }
 
     static public getEnumFromString(s, e) {

lib/WorkflowMain.groovy

@@ -16,7 +16,6 @@ class WorkflowMain {
         def default_invalid = false
 
         // Set defaults common to all run configuration
-
         if (!params.containsKey('genome_version')) {
             if (Constants.GENOMES_VERSION_37.contains(params.genome)) {
                 params.genome_version = '37'
@@ -200,7 +199,7 @@ class WorkflowMain {
                 def panels = Constants.PANELS_DEFINED.join('\n    - ')
                 log.error "\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n" +
                     "  A panel is required to be set using the --panel CLI argument or in a\n" +
-                    "  configuration file when running in targeted mode.\n" +
+                    "  configuration file when running in targeted mode or panel resource creation mode.\n" +
                     "  Currently, the available built-in panels are:\n" +
                     "    - ${panels}\n" +
                     "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~"

main.nf

@@ -58,8 +58,9 @@ if (workflow.stubRun && params.create_stub_placeholders) {
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 */
 
-include { TARGETED } from './workflows/targeted'
-include { WGTS     } from './workflows/wgts'
+include { PANEL_RESOURCE_CREATION } from './workflows/panel_resource_creation'
+include { TARGETED                } from './workflows/targeted'
+include { WGTS                    } from './workflows/wgts'
 
 /*
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -78,6 +79,8 @@ workflow NFCORE_ONCOANALYSER {
         WGTS()
     } else if (run_mode === Constants.RunMode.TARGETED) {
         TARGETED()
+    } else if (run_mode === Constants.RunMode.PANEL_RESOURCE_CREATION) {
+        PANEL_RESOURCE_CREATION()
     } else {
         log.error("received bad run mode: ${run_mode}")
         Nextflow.exit(1)

modules/local/amber/main.nf

@@ -11,7 +11,7 @@ process AMBER {
     tuple val(meta), path(tumor_bam), path(normal_bam), path(donor_bam), path(tumor_bai), path(normal_bai), path(donor_bai)
     val genome_ver
     path heterozygous_sites
-    path target_region_bed
+    path target_regions_bed
 
     output:
     tuple val(meta), path('amber/'), emit: amber_dir
@@ -35,7 +35,7 @@ process AMBER {
     if (donor_bam) reference_bams.add(donor_bam.toString())
     def reference_bam_arg = reference_bams.size() > 0 ? "-reference_bam ${String.join(",", reference_bams)}" : ''
 
-    def target_regions_bed_arg = target_region_bed ? "-target_regions_bed ${target_region_bed}" : ''
+    def target_regions_bed_arg = target_regions_bed ? "-target_regions_bed ${target_regions_bed}" : ''
 
     """
     amber \\

modules/local/amber/meta.yml

@@ -46,19 +46,15 @@ input:
       type: file
       description: AMBER heterozygous sites file
       pattern: "*.{vcf.gz}"
-  - target_region_bed:
+  - target_regions_bed:
       type: file
-      description: Target region BED file (optional)
+      description: Target regions BED file (optional)
       pattern: "*.{bed}"
 output:
-  - meta:
-      type: map
-      description: |
-        Groovy Map containing sample information
-        e.g. [id: 'sample_id', tumor_id: 'tumor_name', normal_id: 'normal_name']
-  - amber_dir:
-      type: directory
+  - cobalt_normalisation:
+      file: directory
       description: AMBER output directory
+      pattern: "versions.yml"
   - versions:
       type: file
       description: File containing software versions

modules/local/cobalt/panel_normalisation/environment.yml

@@ -0,0 +1,7 @@
+name: cobalt_panel_normalisation
+channels:
+  - conda-forge
+  - bioconda
+  - defaults
+dependencies:
+  - bioconda::hmftools-cobalt=2.0

modules/local/cobalt/panel_normalisation/main.nf

@@ -0,0 +1,62 @@
+process COBALT_PANEL_NORMALISATION {
+    label 'process_single'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/hmftools-cobalt:2.0--hdfd78af_0' :
+        'biocontainers/hmftools-cobalt:2.0--hdfd78af_0' }"
+
+    input:
+    tuple path('amber_dir.*'), path('cobalt_dir.*')
+    val genome_ver
+    path gc_profile
+    path target_regions_bed
+
+    output:
+    path 'cobalt.region_normalisation.*.tsv', emit: cobalt_normalisation
+    path 'versions.yml'                     , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+
+    """
+    mkdir -p inputs/
+
+    for fp in \$(find -L amber_dir.* cobalt_dir.* -type f ! -name '*.version'); do
+        ln -sf ../\${fp} inputs/\${fp##*/};
+    done
+
+    (
+        echo SampleId;
+        find -L inputs/ -type f -name '*.amber.baf.tsv.gz' | sed 's#inputs/##; s#\\.amber\\..*\$##; s#\\.cobalt\\..*\$##' | sort -V | uniq;
+    ) > sample_ids.txt
+
+    java -cp /usr/local/share/hmftools-cobalt-2.0-0/cobalt.jar \\
+        -Xmx${Math.round(task.memory.bytes * 0.95)} \\
+        com.hartwig.hmftools.cobalt.norm.NormalisationFileBuilder \\
+        ${args} \\
+        -sample_id_file sample_ids.txt \\
+        -amber_dir inputs/ \\
+        -cobalt_dir inputs/ \\
+        -ref_genome_version ${genome_ver} \\
+        -gc_profile ${gc_profile} \\
+        -target_regions_bed ${target_regions_bed} \\
+        -output_file cobalt.region_normalisation.${genome_ver}.tsv \\
+        -log_level ${params.module_log_level}
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        cobalt_panel_normalisation: \$(cobalt -version | sed 's/^.* //')
+    END_VERSIONS
+    """
+
+    stub:
+    """
+    touch cobalt.region_normalisation.${genome_ver}.tsv
+
+    echo -e '${task.process}:\\n  stub: noversions\\n' > versions.yml
+    """
+}

modules/local/cobalt/panel_normalisation/meta.yml

@@ -0,0 +1,41 @@
+name: cobalt_panel_normalisation
+description: Count bam lines determines the read depth ratios of the supplied tumor and reference genomes
+keywords:
+  - cobalt
+  - read depth ratios
+  - cnv
+tools:
+  - cobalt:
+      description: Count bam lines determines the read depth ratios of the supplied tumor and reference genomes.
+      homepage: https://github.com/hartwigmedical/hmftools/tree/master/cobalt
+      documentation: https://github.com/hartwigmedical/hmftools/tree/master/cobalt
+      licence: ["GPL v3"]
+input:
+  - amber_dirs:
+      type: directory
+      description: List of AMBER output directories
+  - cobalt_dirs:
+      type: directory
+      description: List of COBALT output directories
+  - genome_ver:
+      type: string
+      description: Reference genome version
+  - gc_profile:
+      type: file
+      description: GC profile file
+      pattern: "*.{cnp}"
+  - target_regions_bed:
+      type: file
+      description: Target regions BED file
+      pattern: "*.{bed}"
+output:
+  - cobalt_normalisation:
+      type: file
+      description: COBALT normalisation file
+      pattern: "*.{tsv}"
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+authors:
+  - "@scwatts"

modules/local/cobalt/environment.yml → modules/local/cobalt/run/environment.yml

@@ -1,4 +1,4 @@
-name: cobalt
+name: cobalt_run
 channels:
   - conda-forge
   - bioconda

modules/local/cobalt/main.nf → modules/local/cobalt/run/main.nf

@@ -52,7 +52,7 @@ process COBALT {
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":
-        cobalt: \$(cobalt -version | sed -n '/^Cobalt version/ { s/^.* //p }')
+        cobalt_run: \$(cobalt -version | sed -n '/^Cobalt version/ { s/^.* //p }')
     END_VERSIONS
     """
 

modules/local/cobalt/meta.yml → modules/local/cobalt/run/meta.yml

@@ -1,4 +1,4 @@
-name: cobalt
+name: cobalt_run
 description: Count bam lines determines the read depth ratios of the supplied tumor and reference genomes
 keywords:
   - cobalt

modules/local/isofox/panel_normalisation/environment.yml

@@ -0,0 +1,7 @@
+name: isofox_panel_normalisation
+channels:
+  - conda-forge
+  - bioconda
+  - defaults
+dependencies:
+  - bioconda::hmftools-isofox=1.7.1

modules/local/isofox/panel_normalisation/main.nf

@@ -0,0 +1,60 @@
+process ISOFOX_PANEL_NORMALISATION {
+    label 'process_medium'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/hmftools-isofox:1.7.1--hdfd78af_0' :
+        'biocontainers/hmftools-isofox:1.7.1--hdfd78af_0' }"
+
+    input:
+    path 'isofox_dirs.*'
+    val genome_ver
+    path gene_ids
+    path gene_distribution
+
+    output:
+    path 'isofox.gene_normalisation.*.csv', emit: isofox_normalisation
+    path 'versions.yml'                   , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+
+    """
+    mkdir -p inputs/
+    for fp in \$(find -L isofox_dirs.* -name '*.gene_data.csv'); do ln -sf ../\${fp} inputs/; done
+
+    (
+       echo SampleId;
+       find inputs/ -name '*csv' | sed 's#^.*/\\(.*\\).isf.gene_data.csv#\\1#';
+    ) > sample_ids.txt
+
+    java -cp /usr/local/share/hmftools-isofox-1.7.1-0/isofox.jar \\
+        -Xmx${Math.round(task.memory.bytes * 0.95)} \\
+        com.hartwig.hmftools.isofox.cohort.CohortAnalyser \\
+        ${args} \\
+        -sample_data_file sample_ids.txt \\
+        -root_data_dir inputs/ \\
+        -analyses PANEL_TPM_NORMALISATION \\
+        -gene_id_file ${gene_ids} \\
+        -gene_distribution_file ${gene_distribution} \\
+        -output_dir ./ \\
+        -log_level ${params.module_log_level}
+
+    mv isofox.panel_gene_normalisation.csv isofox.gene_normalisation.${genome_ver}.csv
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        isofox: \$(isofox -version | sed 's/^.* //')
+    END_VERSIONS
+    """
+
+    stub:
+    """
+    touch isofox.gene_normalisation.${genome_ver}.csv
+
+    echo -e '${task.process}:\\n  stub: noversions\\n' > versions.yml
+    """
+}

modules/local/isofox/panel_normalisation/meta.yml

@@ -0,0 +1,37 @@
+name: isofox_panel_normalisation
+description: Characterise and count gene, transcript features
+keywords:
+  - rna
+  - rnaseq
+tools:
+  - isofox:
+      description: Characterises and counts gene, transcript features
+      homepage: https://github.com/hartwigmedical/hmftools/tree/master/isofox
+      documentation: https://github.com/hartwigmedical/hmftools/tree/master/isofox
+      licence: ["GPL v3"]
+input:
+  - isofox_dirs:
+      type: directory
+      description: List of Isofox directories
+  - genome_ver:
+      type: string
+      description: Reference genome version
+  - gene_ids:
+      type: file
+      description: Isofox gene ID file (optional)
+      pattern: "*.{csv}"
+  - gene_distribution:
+      type: file
+      description: Isofox cohort gene expression file
+      pattern: "*.{csv}"
+output:
+  - isofox_normalisation:
+      type: file
+      description: Isofox normalisation file
+      pattern: "versions.yml"
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+authors:
+  - "@scwatts"

modules/local/isofox/environment.yml → modules/local/isofox/run/environment.yml

@@ -1,4 +1,4 @@
-name: isofox
+name: isofox_run
 channels:
   - conda-forge
   - bioconda

modules/local/isofox/meta.yml → modules/local/isofox/run/meta.yml

@@ -1,4 +1,4 @@
-name: isofox
+name: isofox_run
 description: Characterise and count gene, transcript features
 keywords:
   - rna