Merge pull request #151 from icgc-argo-workflows/payload-gen-qc@0.1.0

[release]

Author: lindaxiang

payload-gen-qc/.dockerignore

@@ -0,0 +1,5 @@
+.gitignore
+.nextflow*
+tests
+work
+outdir

payload-gen-qc/Dockerfile

@@ -0,0 +1,20 @@
+FROM python:3.7.5-slim-buster
+
+LABEL org.opencontainers.image.source https://github.com/icgc-argo-workflows/data-processing-utility-tools
+
+RUN apt-get update && apt-get install -y procps
+
+RUN groupadd -g 1000 ubuntu && \
+    useradd -l -u 1000 -g ubuntu ubuntu && \
+    install -d -m 0755 -o ubuntu -g ubuntu /home/ubuntu
+
+ENV PATH="/tools:${PATH}"
+
+COPY *.py /tools/
+
+WORKDIR /tools
+
+USER ubuntu
+
+ENTRYPOINT ["/usr/bin/env"]
+CMD ["/bin/bash"]

payload-gen-qc/README.md

@@ -0,0 +1,49 @@
+# Nextflow Package `payload-gen-qc`
+
+A simple wrapper written in `nextflow` for the payload generation tool to generate ARGO Song payloads containing QC metrics files.  
+
+## Package development
+
+The initial version of this package was created by the WorkFlow Package Manager CLI tool, please refer to
+the [documentation](https://wfpm.readthedocs.io) for details on the development procedure including
+versioning, updating, CI testing and releasing.
+
+
+## Inputs
+### Required
+- `files_to_upload`: All files to upload
+- `metadata_analysis`: JSON file contains donor/sample/specimen/experiment/read_groups/files metadata for input data
+- `wf_name`: Workflow name
+- `wf_version`: Workflow version
+
+### Optional
+- `genome_annotation`: Genome annotation version
+- `genome_build`: Genome build version
+- `cpus`: Set cpu number for running the tool
+- `mem`: Set memory(G) for running the tool
+- `publish_dir`: Specify directory for getting output files
+
+## Outputs
+- `payload`: Payload contains metadata
+- `files_to_upload`: All files to upload with normalized name convention 
+
+## Usage
+
+### Run the package directly
+
+With inputs prepared, you should be able to run the package directly using the following command.
+Please replace the params file with a real one (with all required parameters and input files). Example
+params file(s) can be found in the `tests` folder.
+
+```
+nextflow run icgc-argo-workflows/data-processing-utility-tools/payload-gen-qc/main.nf -r payload-gen-qc.v0.1.0 -params-file <your-params-json-file>
+```
+
+### Import the package as a dependency
+
+To import this package into another package as a dependency, please follow these steps at the
+importing package side:
+
+1. add this package's URI `github.com/icgc-argo-workflows/data-processing-utility-tools/payload-gen-qc@0.1.0` in the `dependencies` list of the `pkg.json` file
+2. run `wfpm install` to install the dependency
+3. add the `include` statement in the main Nextflow script to import the dependent package from this path: `./wfpr_modules/github.com/icgc-argo-workflows/data-processing-utility-tools/payload-gen-qc@0.1.0/main.nf`

payload-gen-qc/main.nf

@@ -0,0 +1,103 @@
+#!/usr/bin/env nextflow
+
+/*
+  Copyright (C) 2021,  Ontario Institute for Cancer Research
+
+  This program is free software: you can redistribute it and/or modify
+  it under the terms of the GNU Affero General Public License as published by
+  the Free Software Foundation, either version 3 of the License, or
+  (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU Affero General Public License for more details.
+
+  You should have received a copy of the GNU Affero General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+  Authors:
+    Linda Xiang
+*/
+
+/********************************************************************/
+/* this block is auto-generated based on info from pkg.json where   */
+/* changes can be made if needed, do NOT modify this block manually */
+nextflow.enable.dsl = 2
+version = '0.1.0'  // package version
+
+container = [
+    'ghcr.io': 'ghcr.io/icgc-argo-workflows/data-processing-utility-tools.payload-gen-qc'
+]
+default_container_registry = 'ghcr.io'
+/********************************************************************/
+
+
+// universal params go here
+params.container_registry = ""
+params.container_version = ""
+params.container = ""
+
+params.cpus = 1
+params.mem = 1  // GB
+params.publish_dir = ""  // set to empty string will disable publishDir
+
+
+// tool specific parmas go here, add / change as needed
+params.files_to_upload = ""
+params.metadata_analysis = ""
+params.wf_name = ""
+params.wf_version = ""
+params.genome_annotation = ""
+params.genome_build = ""
+
+
+process payloadGenQc {
+  container "${params.container ?: container[params.container_registry ?: default_container_registry]}:${params.container_version ?: version}"
+  publishDir "${params.publish_dir}/${task.process.replaceAll(':', '_')}", mode: "copy", enabled: params.publish_dir ? true : false
+
+  cpus params.cpus
+  memory "${params.mem} GB"
+
+  input:  // input, make update as needed
+    path files_to_upload
+    path metadata_analysis
+    val genome_annotation
+    val genome_build
+    val wf_name
+    val wf_version
+
+  output:  // output, make update as needed
+    path "*.payload.json", emit: payload
+    path "out/*", emit: files_to_upload
+
+  script:
+    // add and initialize variables here as needed
+
+    """
+    main.py \
+      -f ${files_to_upload} \
+      -a ${metadata_analysis} \
+      -g "${genome_annotation}" \
+      -b "${genome_build}" \
+      -w "${wf_name}" \
+      -r ${workflow.runName} \
+      -s ${workflow.sessionId} \
+      -v ${wf_version}
+
+    """
+}
+
+
+// this provides an entry point for this main script, so it can be run directly without clone the repo
+// using this command: nextflow run <git_acc>/<repo>/<pkg_name>/<main_script>.nf -r <pkg_name>.v<pkg_version> --params-file xxx
+workflow {
+  payloadGenQc(
+    Channel.fromPath(params.files_to_upload).collect(),
+    file(params.metadata_analysis),
+    params.genome_annotation,
+    params.genome_build,
+    params.wf_name,
+    params.wf_version
+  )
+}