Merge branch 'dev' into remove_squid_pizzly

.github/workflows/ci.yml

@@ -76,3 +76,10 @@ jobs:
           --outdir /home/runner/work/rnafusion/rnafusion/results --starfusion \
           --genomes_base /home/runner/work/rnafusion/rnafusion/results/references \
           --cosmic_username ${{ secrets.COSMIC_USERNAME }} --cosmic_passwd ${{ secrets.COSMIC_PASSWD }}
+
+      - name: Dry test stingtie
+        run: |
+          nextflow run ${GITHUB_WORKSPACE} -profile test,docker -stub \
+          --outdir /home/runner/work/rnafusion/rnafusion/results --stringtie \
+          --genomes_base /home/runner/work/rnafusion/rnafusion/results/references \
+          --cosmic_username ${{ secrets.COSMIC_USERNAME }} --cosmic_passwd ${{ secrets.COSMIC_PASSWD }}

CHANGELOG.md

@@ -7,13 +7,25 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Added
 
+- Add picard CollectInserSizeMetrics to QC workflow [#408](https://github.com/nf-core/rnafusion/pull/408)
+
 ### Changed
 
+- Replace PICARD_MARKDUPLICATES with GATK4_MARKDUPLICATES [#409](https://github.com/nf-core/rnafusion/pull/409)
+- Removed `--fusioninspector_filter` and `--fusionreport_filter` in favor of `--tools_cutoff` (default = 1, no filters applied) [#389](https://github.com/nf-core/rnafusion/pull/389)
+- Now publishing convert2bed output to convert2bed to keep the output file for mosdepth [#420](https://github.com/nf-core/rnafusion/pull/420)
+- No more checks for existence of samplesheet, which made building references fail (building references uses a fake sample sheet if none is provided) [#420](https://github.com/nf-core/rnafusion/pull/420)
+- `--extreme_sensitivity` used for fusioninspector to minimize fusioninspector filtering [#424](https://github.com/nf-core/rnafusion/pull/424)
+
 ### Fixed
 
+- Fix channel i/o issue in StringTie workflow and add StringTie in github CI tests [#416](https://github.com/nf-core/rnafusion/pull/416)
+- Updated COSMIC database to fix 404 error while downloading fusionreport references [#420](https://github.com/nf-core/rnafusion/pull/420)
+
 ### Removed
 
 - Remove `squid` and `pizzly` fusion detection tools [#406](https://github.com/nf-core/rnafusion/pull/406)
+- Remove harsh trimming option `--trim` [#413](https://github.com/nf-core/rnafusion/pull/413)
 
 ## v2.4.0 - [2023/09/22]
 

README.md

@@ -56,7 +56,7 @@ In rnafusion the full-sized test includes reference building and fusion detectio
     - [FusionInspector](https://github.com/FusionInspector/FusionInspector)
     - [Arriba](https://github.com/suhrig/arriba) visualisation
     - QC for mapped reads ([`QualiMap: BAM QC`](https://kokonech.github.io/qualimap/HG00096.chr20_bamqc/qualimapReport.html))
-    - Collect metrics ([`picard CollectRnaSeqMetrics`](https://gatk.broadinstitute.org/hc/en-us/articles/360037057492-CollectRnaSeqMetrics-Picard-) and ([`picard MarkDuplicates`](https://gatk.broadinstitute.org/hc/en-us/articles/360037052812-MarkDuplicates-Picard-))
+    - Collect metrics ([`picard CollectRnaSeqMetrics`](https://gatk.broadinstitute.org/hc/en-us/articles/360037057492-CollectRnaSeqMetrics-Picard-), [`picard CollectInsertSizeMetrics`](https://gatk.broadinstitute.org/hc/en-us/articles/360037055772-CollectInsertSizeMetrics-Picard-) and ([`picard MarkDuplicates`](https://gatk.broadinstitute.org/hc/en-us/articles/360037052812-MarkDuplicates-Picard-))
 11. Present QC for raw reads ([`MultiQC`](http://multiqc.info/))
 12. Compress bam files to cram with [samtools view](http://www.htslib.org/)
 

conf/modules.config

@@ -45,6 +45,14 @@ process {
         ]
     }
 
+    withName: CONVERT2BED {
+        publishDir = [
+            path: { "${params.genomes_base}/convert2bed" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
+    }
+
     withName: CUSTOM_DUMPSOFTWAREVERSIONS {
         publishDir = [
             path: { "${params.outdir}/pipeline_info" },
@@ -94,13 +102,12 @@ process {
     withName: FUSIONINSPECTOR {
         ext.when = { !params.skip_vis }
         ext.args = { params.fusioninspector_limitSjdbInsertNsj != 1000000 ? "--STAR_xtra_params \"--limitSjdbInsertNsj ${params.fusioninspector_limitSjdbInsertNsj}\"" : '' }
-
+        ext.args2 = '--extreme_sensitivity'
     }
 
     withName: FUSIONREPORT {
         ext.when         = { !params.skip_vis }
         ext.args         = "--export csv"
-        ext.args2        = { params.fusionreport_filter ? "--tool-cutoff 2" : "--tool-cutoff 1"}
         publishDir = [
             path: { "${params.outdir}/fusionreport/${meta.id}" },
             mode: params.publish_dir_mode,
@@ -162,13 +169,18 @@ process {
         ]
     }
 
-    withName: QUALIMAP_RNASEQ {
-        ext.when         = { !params.skip_qc && !params.fusioninspector_only && (params.starfusion || params.all)}
+    withName: PICARD_COLLECTINSERTSIZEMETRICS {
+        ext.prefix       = { "${meta.id}_collectinsertsize"}
+    }
+        publishDir = [
+            path: { "${params.outdir}/picard" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename },
+        ]
     }
 
-    withName: REFORMAT {
-        ext.args = "forcetrimright=75"
-        ext.args2 = "forcetrimleft=75"
+    withName: QUALIMAP_RNASEQ {
+        ext.when         = { !params.skip_qc && !params.fusioninspector_only && (params.starfusion || params.all)}
     }
 
     withName: SAMPLESHEET_CHECK {

docs/output.md

@@ -270,6 +270,7 @@ Picard CollectRnaMetrics and picard MarkDuplicates share the same output directo
 - `picard`
   - `<sample>.MarkDuplicates.metrics.txt` - metrics from MarkDuplicates
   - `<sample>_rna_metrics.txt` - metrics from CollectRnaMetrics
+  - `<sample>_insert_size_metrics.txt.txt` - metrics from CollectInsertSizeMetrics
   - `<sample>.bam` - BAM file with marked duplicates
 
 </details>

docs/usage.md

@@ -184,10 +184,7 @@ You can also generate such `YAML`/`JSON` files via [nf-core/launch](https://nf-c
 
 #### Trimming
 
-There are 2 options to trim
-
-1. Fastp
-   In this case all tools use the trimmed reads. Quality and adapter trimming by default. In addition, tail trimming and adapter_fastq specification are possible. Example usage:
+When the flag `--fastp_trim` is used, `fastp` is used to provide all tools with trimmed reads. Quality and adapter trimming by default. In addition, tail trimming and adapter_fastq specification are possible. Example usage:
 
 ```bash
 nextflow run nf-core/rnafusion \
@@ -200,18 +197,6 @@ nextflow run nf-core/rnafusion \
 --adapter_fastq <PATH/TO/ADAPTER/FASTQ> (optional)
 ```
 
-2. Hard trimming
-   In this case, only reads fed to fusioncatcher are trimmed. This is a harsh workaround in case of high read-through. The recommended trimming is thus the fastp_trim one. The trimming is done at 75 bp from the tails. Example usage:
-
-```bash
-nextflow run nf-core/rnafusion \
---<tool1> --<tool2> ... \
---input <SAMPLE_SHEET.CSV> \
---genomes_base <PATH/TO/REFERENCES> \
---outdir <OUTPUT/PATH> \
---trim
-```
-
 #### Filter fusions detected by 2 or more tools
 
 ```bash
@@ -220,12 +205,10 @@ nextflow run nf-core/rnafusion \
   --input <SAMPLE_SHEET.CSV> \
   --genomes_base <PATH/TO/REFERENCES> \
   --outdir <OUTPUT/PATH>
-  --fusioninspector_filter
-  --fusionreport_filter
+  --tools_cutoff <INT>
 ```
 
-`--fusioninspector_filter` feed only fusions detected by 2 or more tools to fusioninspector for closer analysis (false by default).
-`--fusionreport_filter` displays only fusions detected by 2 or more tools in fusionreport html index (true by default).
+`--tools_cutoff INT` will discard fusions detected by less than INT tools both for display in fusionreport html index and to consider in fusioninspector.
 
 #### Adding custom fusions to consider as well as the detected set: whitelist
 

modules.json

@@ -45,19 +45,24 @@
                         "git_sha": "cf8f9ace77aac01caa5c7cb92af5bbda7adb77bd",
                         "installed_by": ["modules"]
                     },
+                    "gatk4/markduplicates": {
+                        "branch": "master",
+                        "git_sha": "2aa9c2981930687792ed861b0a5f9ff7bb568a7d",
+                        "installed_by": ["modules"]
+                    },
                     "multiqc": {
                         "branch": "master",
                         "git_sha": "a6e11ac655e744f7ebc724be669dd568ffdc0e80",
                         "installed_by": ["modules"]
                     },
-                    "picard/collectwgsmetrics": {
+                    "picard/collectinsertsizemetrics": {
                         "branch": "master",
-                        "git_sha": "735e1e04e7e01751d2d6e97055bbdb6f70683cc1",
+                        "git_sha": "240937a2a9c30298110753292be041188891f2cb",
                         "installed_by": ["modules"]
                     },
-                    "picard/markduplicates": {
+                    "picard/collectwgsmetrics": {
                         "branch": "master",
-                        "git_sha": "2ee934606f1fdf7fc1cb05d6e8abc13bec8ab448",
+                        "git_sha": "735e1e04e7e01751d2d6e97055bbdb6f70683cc1",
                         "installed_by": ["modules"]
                     },
                     "qualimap/rnaseq": {

modules/local/fusioninspector/main.nf

@@ -21,6 +21,7 @@ process FUSIONINSPECTOR {
     def prefix = task.ext.prefix ?: "${meta.id}"
     def fasta = meta.single_end ? "--left_fq ${reads[0]}" : "--left_fq ${reads[0]} --right_fq ${reads[1]}"
     def args = task.ext.args ?: ''
+    def args2 = task.ext.args2 ?: ''
     """
     FusionInspector \\
         --fusions $fusion_list \\
@@ -29,7 +30,7 @@ process FUSIONINSPECTOR {
         --CPU ${task.cpus} \\
         -O . \\
         --out_prefix $prefix \\
-        --vis $args
+        --vis $args $args2
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

modules/local/fusionreport/detect/main.nf

@@ -4,12 +4,13 @@ process FUSIONREPORT {
 
     // Note: 2.7X indices incompatible with AWS iGenomes.
     conda "bioconda::star=2.7.9a"
-    container "docker.io/clinicalgenomics/fusion-report:2.1.5p4"
+    container "docker.io/clinicalgenomics/fusion-report:2.1.5p5"
 
 
     input:
     tuple val(meta), path(reads), path(arriba_fusions), path(starfusion_fusions),  path(fusioncatcher_fusions)
     tuple val(meta2), path(fusionreport_ref)
+    val(tools_cutoff)
 
     output:
     path "versions.yml"                                                 , emit: versions
@@ -31,7 +32,7 @@ process FUSIONREPORT {
     tools    += params.fusioncatcher  || params.all ? "--fusioncatcher ${fusioncatcher_fusions} " : ''
     def prefix = task.ext.prefix ?: "${meta.id}"
     """
-    fusion_report run $meta.id . $fusionreport_ref $tools --allow-multiple-gene-symbols $args $args2
+    fusion_report run $meta.id . $fusionreport_ref $tools --allow-multiple-gene-symbols --tool-cutoff $tools_cutoff $args $args2
 
     mv fusion_list.tsv ${prefix}.fusionreport.tsv
     mv fusion_list_filtered.tsv ${prefix}.fusionreport_filtered.tsv

modules/local/fusionreport/download/main.nf

@@ -4,7 +4,7 @@ process FUSIONREPORT_DOWNLOAD {
 
     // Note: 2.7X indices incompatible with AWS iGenomes.
     conda "bioconda::star=2.7.9a"
-    container "docker.io/clinicalgenomics/fusion-report:2.1.5p4"
+    container "docker.io/clinicalgenomics/fusion-report:2.1.5p5"
 
     input:
     val(username)