Merge branch 'dev' into 'main'

Dev into master

See merge request tron/easyfuse-pipeline!19

.gitlab-ci.yml

@@ -0,0 +1,7 @@
+easyfuse-pipeline:
+  image: continuumio/miniconda3:23.5.2-0
+  stage: build
+  script:
+    - conda install -c bioconda nextflow=20.10.0
+    - bash test/scripts/test_00.sh
+

CONTRIBUTE.md

@@ -6,3 +6,16 @@
 After the EasyFuse Python package is installed as described in the README, run the integration tests with `make`.
 
 You may need to edit the test scripts to use an appropriate reference folder in `--reference`.
+
+## Prepare a new release
+
+Increase the version in the property `VERSION` in the file `nextflow.config`.
+
+## Test the integration of easyfuse-src
+
+The Python and R code are distributed as a Python package and this is imported in 
+EasyFuse pipeline via conda. But, for development purposes the import of this
+package can be skipped while all other dependencies are still managed via conda.
+
+Use the flag `--disable_pyeasyfuse_conda`, this will rely on whatever version of
+pyeasyfuse has been installed on your environment.

README.md

@@ -15,8 +15,6 @@ For maximal sensitivity, we recommend using an older EasyFuse release with five
  - Publication: [Weber D, Ibn-Salem J, Sorn P, et al. Nat Biotechnol. 2022](https://doi.org/10.1038/s41587-022-01247-9)
 
 
-
-
 ## Usage
 
 ### Dependencies
@@ -37,29 +35,21 @@ tar xvfz easyfuse_ref_v2.tar.gz
 ```
 
 
-### Download nextflow pipeline and install EasyFuse package
+### Install the nextflow pipeline
 
-Next, you have to download the nextflow pipeline including the EasyFuse package source dir.
+There are two alternatives, manually install the workflow or let Nexftlow handle this via the GitHub repository.
 
+To install manually:
 ```
-git clone --recurse-submodules https://github.com/TRON-Bioinformatics/EasyFuse.git
-
+git clone https://github.com/TRON-Bioinformatics/EasyFuse.git
 cd EasyFuse
+```
 
-cd easyfuse_src
-
-# Create virtual environment using Python3.7 (only works on Python3.7)
-python3.7 -m venv env/
-
-# Activate environment
-source env/bin/activate
-
-# Install poetry for easy installation
-pip install poetry
-
-# Install EasyFuse package using poetry
-poetry install
+To install with Nextflow (only available from release 2.0.1):
 ```
+nextflow run tron-bioinformatics/easyfuse -r x.y.z --help
+```
+where x.y.z corresponds to an EasyFuse release.
 
 
 ### Run the pipeline
@@ -74,11 +64,21 @@ sample_01	/path/to/sample_01_R1.fastq.gz	/path/to/sample_01_R2.fastq.gz
 ```
 
 
-Start the pipeline as follows
+Start the pipeline as follows if you installed manually
 
 ```
 nextflow main.nf \
-  -profile conda \
+  -profile conda -with-conda \
+  --reference /path/to/reference/folder \
+  --input_files /path/to/input_table_file \
+  --output /path/to/output_folder
+```
+
+
+Or as follows if you installed it via Nextflow (only available from release 2.0.1):
+```
+nextflow run tron-bioinformatics/easyfuse -r x.y.z \
+  -profile conda -with-conda \
   --reference /path/to/reference/folder \
   --input_files /path/to/input_table_file \
   --output /path/to/output_folder

environments/easyfuse_src.yml

@@ -0,0 +1,8 @@
+name: easyfuse_src
+channels:
+  - conda-forge
+  - bioconda
+  - defaults
+dependencies:
+  - bioconda::pyeasyfuse=2.0.3
+  - bioconda::skewer=0.2.2

environments/qc.yml

@@ -5,4 +5,3 @@ channels:
   - defaults
 dependencies:
   - bioconda::fastqc=0.11.9
-  - bioconda::skewer=0.2.2

environments/requantification.yml

@@ -5,4 +5,4 @@ channels:
   - defaults
 dependencies:
   - bioconda::star=2.6.1d
-  - bioconda::samtools=1.9.0
+  - bioconda::pyeasyfuse=2.0.3

environments/requantification_wo_easyfuse.yml

@@ -0,0 +1,8 @@
+name: requantification
+channels:
+  - conda-forge
+  - bioconda
+  - defaults
+dependencies:
+  - bioconda::star=2.6.1d
+  - bioconda::samtools=1.9.0

modules/01_qc.nf

@@ -28,7 +28,7 @@ process FASTQC_PARSER {
     memory "8g"
     tag "${name}"
 
-//    conda (params.enable_conda ? "bioconda::fastqc=0.11.9" : null)
+    conda (params.enable_conda && ! params.disable_pyeasyfuse_conda ? "${baseDir}/environments/easyfuse_src.yml" : null)
 
     input:
     tuple val(name),
@@ -40,7 +40,7 @@ process FASTQC_PARSER {
         path("qc_table.txt"), emit: qc_table
 
     """
-    easy-fuse qc-parser -i $fastq1_qc_data $fastq1_qc_data -o qc_table.txt
+    easy-fuse qc-parser -i $fastq1_qc_data $fastq2_qc_data -o qc_table.txt
     """
 }
 
@@ -49,7 +49,7 @@ process SKEWER {
     memory "8g"
     tag "${name}"
 
-    conda (params.enable_conda ? "${baseDir}/environments/qc.yml" : null)
+    conda (params.enable_conda ? "${baseDir}/environments/easyfuse_src.yml" : null)
 
     input:
     tuple val(name),

modules/02_alignment.nf

@@ -45,7 +45,7 @@ process READ_FILTER {
     memory "8g"
     tag "${name}"
 
-    //conda (params.enable_conda ? "bioconda::easyfuse=0.1.0" : null)
+    conda (params.enable_conda && ! params.disable_pyeasyfuse_conda ? "${baseDir}/environments/easyfuse_src.yml" : null)
 
     input:
       tuple val(name), path(bam)

modules/04_joint_fusion_calling.nf

@@ -4,7 +4,7 @@ process FUSION_PARSER {
     memory "8g"
     tag "${name}"
 
-    //conda (params.enable_conda ? "bioconda::easyfuse=0.1.0" : null)
+    conda (params.enable_conda && ! params.disable_pyeasyfuse_conda ? "${baseDir}/environments/easyfuse_src.yml" : null)
 
     input:
       tuple val(name), path(fusion_catcher_1), path(fusion_catcher_2), path(star_fusion)
@@ -30,7 +30,7 @@ process FUSION_ANNOTATION {
     memory "8g"
     tag "${name}"
 
-    //conda (params.enable_conda ? "bioconda::easyfuse=0.1.0" : null)
+    conda (params.enable_conda && ! params.disable_pyeasyfuse_conda ? "${baseDir}/environments/easyfuse_src.yml" : null)
 
     input:
       tuple val(name), path(fusions)

modules/05_requantification.nf

@@ -4,8 +4,13 @@ process STAR_INDEX {
     memory "8g"
     tag "${name}"
 
-    //conda (params.enable_conda ? "bioconda::easyfuse=0.1.0" : null)
-    conda (params.enable_conda ? "${baseDir}/environments/requantification.yml" : null)
+    // NOTE: we really need STAR and samtools here, these are not added as dependencies to the bioconda package
+    if (params.disable_pyeasyfuse_conda) {
+        conda (params.enable_conda ? "${baseDir}/environments/requantification_wo_easyfuse.yml" : null)
+    }
+    else {
+        conda (params.enable_conda ? "${baseDir}/environments/requantification.yml" : null)
+    }
 
     input:
       tuple val(name), path(annot_csv), path(annot_fasta)
@@ -30,7 +35,7 @@ process FUSION_FILTER {
     memory "8g"
     tag "${name}"
 
-    //conda (params.enable_conda ? "bioconda::easyfuse=0.1.0" : null)
+    conda (params.enable_conda && ! params.disable_pyeasyfuse_conda ? "${baseDir}/environments/easyfuse_src.yml" : null)
 
     input:
       tuple val(name), path(bam), path(annot_fusions_1), path(annot_fusions_2), path(read_stats)
@@ -54,7 +59,7 @@ process STAR_CUSTOM {
     memory "32g"
     tag "${name}"
 
-    conda (params.enable_conda ? "${baseDir}/environments/requantification.yml" : null)
+    conda (params.enable_conda ? "${baseDir}/environments/requantification_wo_easyfuse.yml" : null)
 
     input:
       tuple val(name), path(fastq1), file(fastq2), path(star_index)
@@ -87,7 +92,8 @@ process READ_COUNT {
   cpus 6
   memory "50g"
   tag "${name}"
-  //publishDir "${params.output}/${name}", mode: 'copy'
+
+  conda (params.enable_conda && ! params.disable_pyeasyfuse_conda ? "${baseDir}/environments/easyfuse_src.yml" : null)
 
   input:
     tuple val(name), path(bam), path(bam_index), path(read_stats)

modules/06_summarize.nf

@@ -5,7 +5,7 @@ process SUMMARIZE_DATA {
     tag "${name}"
     publishDir "${params.output}/${name}", mode: 'copy'
 
-    conda (params.enable_conda ? "${baseDir}/environments/summarize_data.yml" : null)
+    conda (params.enable_conda && ! params.disable_pyeasyfuse_conda ? "${baseDir}/environments/easyfuse_src.yml" : null)
 
     input:
       tuple val(name), path(fusions), path(annot_csv), path(annot_fasta), path(cpms), path(counts), path(read_stats)

nextflow.config

@@ -17,7 +17,7 @@ env {
 // Capture exit codes from upstream processes when piping
 process.shell = ['/bin/bash', '-euo', 'pipefail']
 
-VERSION = '2.0.0a1'
+VERSION = '2.0.1'
 
 manifest {
   name = 'TRON-Bioinformatics/easy-fuse-workflow'
@@ -43,6 +43,9 @@ params.annotation_db = "${params.reference}/Homo_sapiens.GRCh38.86.gff3.db"
 params.reference_tsl = "${params.reference}/Homo_sapiens.GRCh38.86.gtf.tsl"
 params.model_pred = "${baseDir}/data/model/Fusion_modeling_FFPE_train_v35.random_forest.model_full_data.EF_requant_type.rds"
 
+// use this to avoid
+params.disable_pyeasyfuse_conda = false
+
 
 
 params.help_message = """