Merge pull request #62 from mskcc/feature/agfusion_module

Feature/agfusion module

conf/modules.config

@@ -179,9 +179,21 @@ process {
             pattern: '*.tsv'
         ]
     }
+
+    withName: AGFUSION_BATCH {
+        ext.args = "-a cff_ext --noncanonical --middlestar"
+        publishDir = [
+            path: { "${params.outdir}/analysis/${meta.id}/agfusion" },
+            mode: params.publish_dir_mode,
+            pattern: 'fusion_transcripts.csv',
+            enabled: false
+        ]
+    }
+
     withName: MERGE_CFF {
         ext.args = {  "-H " }
     }
+
     withName: ONCOKB_FUSIONANNOTATOR {
         ext.when = params.run_oncokb_fusionannotator
         secret = ["ONCOKB_TOKEN"]

modules/local/agfusion/batch/main.nf

@@ -0,0 +1,43 @@
+process AGFUSION_BATCH {
+    tag "$meta.id"
+    label 'process_low'
+
+    // Note: 2.7X indices incompatible with AWS iGenomes.
+    conda 'bioconda::agfusion=1.252'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'cmopipeline/agfusion:0.0.3' :
+        'cmopipeline/agfusion:0.0.3' }"
+
+    input:
+    tuple val(meta), path(fusions)
+    path(agfusion_db)
+    path(pyensembl_cache)
+
+    output:
+    tuple val(meta), path("${prefix}")                       , emit: fusions_annotated
+    tuple val(meta), path("${prefix}.fusion_transcripts.csv"), emit: fusion_transcripts
+    path "versions.yml"                                      , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args   = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    export PYENSEMBL_CACHE_DIR=\$PWD/${pyensembl_cache}
+
+    agfusion batch \\
+        -f ${fusions} \\
+        -db ${agfusion_db} \\
+        -o ${prefix} \\
+        ${args}
+
+    cat ${prefix}/*/*.fusion_transcripts.csv | awk -F"," -v OFS="\\t" 'NR != 1 && FNR == 1 {next;}{print}' > fusion_transcripts.csv
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        agfusion: \$(agfusion -v) (fork)
+    END_VERSIONS
+    """
+}

modules/local/agfusion/batch/meta.yml

@@ -0,0 +1,49 @@
+name: agfusion_batch
+description: AGFusion batch annotation
+keywords:
+  - agfusion
+  - fusion
+  - batch
+  - frame
+tools:
+  - batch:
+      description: Annotate a file containing fusions
+      homepage: https://github.com/anoronh4/AGFusion
+      documentation: https://github.com/anoronh4/AGFusion/blob/master/README.md
+
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - fusions:
+      type: path
+      description: Fusions in one of many formats (may include arriba outputs or cff)
+      pattern: "*"
+  - agfusion_db:
+      type: path
+      description: File containing AGFusion reference information.
+      pattern: "*.db"
+  - pyensembl_cache:
+      type: path
+      description: Folder containing pyensembl cache.
+      pattern: "*"
+
+output:
+  - versions:
+      type: path
+      description: File containing software versions
+      pattern: "versions.yml"
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - agfusion_result:
+      type: path
+      description: Folder containing annotations
+      pattern: "${prefix}/"
+
+authors:
+  - "@anoronh4"

modules/local/agfusion/container/Dockerfile

@@ -0,0 +1,26 @@
+FROM ubuntu:bionic-20230530
+
+LABEL maintainer="Anne Marie Noronha ([email protected])" \
+    version.image="0.0.3"
+
+# INSTALL DEPENDENCIES
+
+ENV DEBIAN_FRONTEND=noninteractive
+
+RUN apt-get update -y
+RUN apt-get install -y build-essential python3 python3-pip python3-matplotlib python3-pandas python3-future python3-biopython curl less vim libnss-sss git zip
+RUN pip3 install --upgrade pip
+RUN pip3 install pyensembl
+
+# Additional libraries needed for AGFusion build command
+RUN apt-get install -y default-libmysqlclient-dev
+RUN pip3 install mysqlclient
+
+# INSTALL AGFUSION & DATABASE FILES
+WORKDIR /usr/local/bin
+RUN git clone https://github.com/anoronh4/AGFusion.git --branch feature/metafusion_parser
+WORKDIR /usr/local/bin/AGFusion
+RUN pip3 install .
+
+# downgrade pyensembl for compatibility
+RUN pip3 install gtfparse==1.2.1 --upgrade

modules/local/agfusion/download/main.nf

@@ -0,0 +1,58 @@
+process AGFUSION_DOWNLOAD {
+    label 'process_low'
+
+    // Note: 2.7X indices incompatible with AWS iGenomes.
+    conda 'bioconda::agfusion=1.252'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'cmopipeline/agfusion:0.0.3' :
+        'cmopipeline/agfusion:0.0.3' }"
+
+    input:
+    val(ensembl_release)
+    val(genome)
+
+    output:
+    path "agfusion.*.db"  , emit: agfusion_db
+    path "pyensembl_cache", emit: pyensembl_cache
+    path "versions.yml"   , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def agfusion_genome = ['GRCh37','smallGRCh37','hg19'].contains(genome) ? 'hg19' :
+        ['GRCh38','hg38'].contains(genome) ? 'hg38' :
+        ['GRCm38','mm10'].contains(genome) ? 'mm10' :
+    def pyensembl_species = ['GRCm38','mm10'].contains(genome) ? 'mus_musculus' : 'homo_sapiens'
+    if (ensembl_release < 93) {
+        """
+        export PYENSEMBL_CACHE_DIR=\$PWD/pyensembl_cache
+
+        pyensembl install --species ${pyensembl_species} --release ${ensembl_release}
+
+        agfusion download -g ${agfusion_genome}
+
+        cat <<-END_VERSIONS > versions.yml
+        "${task.process}":
+            agfusion: \$(agfusion -v) (fork)
+        END_VERSIONS
+        """
+    } else {
+        """
+        export PYENSEMBL_CACHE_DIR=\$PWD/pyensembl_cache
+
+        pyensembl install --species ${pyensembl_species} --release ${ensembl_release}
+
+        curl http://ftp.ebi.ac.uk/pub/databases/Pfam/current_release/database_files/pfamA.txt.gz > pfamA.txt.gz
+        gunzip pfamA.txt.gz
+        agfusion build --dir . --species ${agfusion_genome} --release ${ensembl_release} --pfam pfamA.txt
+        rm pfamA.txt
+
+        cat <<-END_VERSIONS > versions.yml
+        "${task.process}":
+            agfusion: \$(agfusion -v) (fork)
+        END_VERSIONS
+        """
+    }
+}

modules/local/agfusion/download/meta.yml

@@ -0,0 +1,35 @@
+name: fusionreport_download
+description: Build DB for fusionreport
+keywords:
+  - sort
+tools:
+  - fusioncatcher:
+      description: Build DB for fusionreport
+      homepage: https://github.com/ndaniel/fusioncatcher/
+      documentation: https://github.com/ndaniel/fusioncatcher/blob/master/doc/manual.md
+      tool_dev_url: https://github.com/ndaniel/fusioncatcher/
+      doi: "10.1101/011650"
+      licence: ["GPL v3"]
+
+input:
+  - username:
+      type: value
+      description: Organism for which the data is downloaded from Ensembl database and built
+      pattern: "*"
+  - passwd:
+      type: value
+      description: Organism for which the data is downloaded from Ensembl database and built
+      pattern: "*"
+
+output:
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+  - reference:
+      type: directory
+      description: directory containing the genome resource files required for fusioncatcher
+      pattern: "fusioncatcher-genome"
+
+authors:
+  - "@praveenraj2018"