Feature/fusion vcf

conf/modules.config

@@ -46,6 +46,22 @@ process {
         time   = { 10.m + (1.ms * reads.sum { it.size() } / reads.count { it } / 1000) * task.attempt }
     }
 
+    withName: FUSIONINSPECTOR {
+        cpus   = { 12 }
+        memory = { 40.GB * task.attempt }
+        time   = { 4.h * task.attempt }
+
+
+        ext.args = { params.fusioninspector_limitSjdbInsertNsj != 1000000 ? "--STAR_xtra_params \"--limitSjdbInsertNsj ${params.fusioninspector_limitSjdbInsertNsj}\"" : '' }
+        ext.args2 = '--annotate --examine_coding_effect'
+        publishDir = [
+            path: { "${params.outdir}/fusions/fusioninspector/${meta.id}" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
+    }
+
+
     withName: MULTIQC {
         clusterOptions = "${params.cluster_options}"
         ext.args       = { params.multiqc_title ? "--title \"${params.multiqc_title}\"" : "--title \"${params.analysis_id}\"" }
@@ -230,6 +246,7 @@ process {
     }
     withName: FUSIONREPORT_DETECT {
         ext.args = {'--no-cosmic'}
+        ext.args2 = "--export csv"
 
         publishDir     = [
             path: { "${params.outdir}/fusions/fusion_report/${meta.id}/" },
@@ -395,4 +412,26 @@ process {
         ]
     }
 
+
+    withName: HGNC_DOWNLOAD {
+        cpus = 1
+        memory = { 1.GB }
+        time = { 5.m * task.attempt }
+    }
+
+    withName: VCF_COLLECT {
+        cpus = 2
+        memory = { 5.GB }
+        time = { 10.m * task.attempt }
+
+        publishDir = [
+            [
+                path: "${params.outdir}/fusions/vcf/",
+                mode: params.publish_dir_mode,
+                saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+            ]
+        ]
+    }
+
+
 }

modules.json

@@ -5,6 +5,11 @@
         "https://github.com/nf-core/modules.git": {
             "modules": {
                 "nf-core": {
+                    "agat/convertspgff2tsv": {
+                        "branch": "master",
+                        "git_sha": "55ed5f4aabcdcd7a4cc44e1700f956756c7e8532",
+                        "installed_by": ["modules"]
+                    },
                     "arriba/arriba": {
                         "branch": "master",
                         "git_sha": "025a6f75e1f72e4ab60abb4bd65b3f289d4ad910",
@@ -15,6 +20,11 @@
                         "git_sha": "dc94b6ee04a05ddb9f7ae050712ff30a13149164",
                         "installed_by": ["modules"]
                     },
+                    "fusioninspector": {
+                        "branch": "master",
+                        "git_sha": "310a7a59c7f2362d25070e5928f3139f92377eaf",
+                        "installed_by": ["modules"]
+                    },
                     "fusionreport/detect": {
                         "branch": "master",
                         "git_sha": "e753770db613ce014b3c4bc94f6cba443427b726",

modules/local/hgnc_download/main.nf

@@ -0,0 +1,38 @@
+process HGNC_DOWNLOAD {
+    tag "hgnc"
+    label 'process_low'
+
+    conda "bioconda::gnu-wget=1.18"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://community-cr-prod.seqera.io/docker/registry/v2/blobs/sha256/3b/3b54fa9135194c72a18d00db6b399c03248103f87e43ca75e4b50d61179994b3/data' :
+        'community.wave.seqera.io/library/wget:1.21.4--8b0fcde81c17be5e' }"
+
+    output:
+    path "hgnc_complete_set.txt"        , emit: hgnc_ref
+    path "HGNC-DB-timestamp.txt"        , emit: hgnc_date
+    path "versions.yml"                 , emit: versions
+
+
+    script:
+    """
+    wget --no-check-certificate https://storage.googleapis.com/public-download-files/hgnc/tsv/tsv/hgnc_complete_set.txt
+    date +%Y-%m-%d/%H:%M  > HGNC-DB-timestamp.txt
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        wget: \$(wget --version | head -1 | cut -d ' ' -f 3)
+    END_VERSIONS
+    """
+
+    stub:
+    """
+    touch "hgnc_complete_set.txt"
+    touch "HGNC-DB-timestamp.txt"
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        wget: \$(wget --version | head -1 | cut -d ' ' -f 3)
+    END_VERSIONS
+    """
+
+}

modules/local/vcf_collect/main.nf

@@ -0,0 +1,46 @@
+process VCF_COLLECT {
+    tag "$meta.id"
+    label 'process_single'
+
+    conda "conda-forge::pandas=1.5.2"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/pandas:1.5.2' :
+        'biocontainers/pandas:1.5.2' }"
+
+    input:
+    tuple val(meta), path(fusioninspector_tsv), path(fusioninspector_gtf_tsv), path(fusionreport_report), path(fusionreport_csv)
+    tuple val(meta2),  path(hgnc_ref)
+    tuple val(meta3),  path(hgnc_date)
+
+    output:
+    path "versions.yml"              , emit: versions
+    tuple val(meta), path("*vcf")    , emit: vcf
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    vcf_collect.py --fusioninspector $fusioninspector_tsv --fusionreport $fusionreport_report --fusioninspector_gtf $fusioninspector_gtf_tsv --fusionreport_csv $fusionreport_csv --hgnc $hgnc_ref --sample ${prefix} --out ${prefix}_fusion_data.vcf
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        python: \$(python --version | sed 's/Python //g')
+        vcf_collect: 0.1
+        HGNC DB retrieval: \$(cat $hgnc_date)
+    END_VERSIONS
+    """
+
+    stub:
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch ${prefix}_fusion_data.vcf
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        python: \$(python --version | sed 's/Python //g')
+        vcf_collect: 0.1
+    END_VERSIONS
+    """
+}