Merge pull request #121 from heylf/dev

Dev - Adding FastQC to the pipeline, with integration into MultiQC

docs/output.md

@@ -12,6 +12,7 @@ The pipeline is built using [Nextflow](https://www.nextflow.io/) and processes d
  - [:warning: Please read this documentation on the nf-core website: https://nf-co.re/scrnaseq/output](#warning-please-read-this-documentation-on-the-nf-core-website-httpsnf-corescrnaseqoutput)
  - [Introduction](#introduction)
  - [Pipeline overview](#pipeline-overview)
+ - [FastQC](#fastqc)
  - [Kallisto & Bustools Results](#kallisto--bustools-results)
  - [STARsolo](#starsolo)
  - [Salmon Alevin & AlevinQC](#salmon-alevin--alevinqc)
@@ -20,6 +21,19 @@ The pipeline is built using [Nextflow](https://www.nextflow.io/) and processes d
  - [MultiQC](#multiqc)
  - [Pipeline information](#pipeline-information)
 
+## FastQC
+
+See [FastQC](https://www.bioinformatics.babraham.ac.uk/projects/fastqc) for details about FastQC.
+
+The pipeline analyzes the raw data and generates for each file a FastQC report. All report are collected in MultiQC.
+
+**Output directory: `results/fastqc`**
+
+- `.html`
+ - Contains the FastQC report.
+- `.zip`
+ - Contains additional information, such as individual plots, and FastQC raw data.
+
 ## Kallisto & Bustools Results
 
 See [Kallisto](https://pachterlab.github.io/kallisto/about) for details about Kallisto and [Bustools](https://bustools.github.io/) for more information on BusTools.

modules/local/multiqc.nf

@@ -1,30 +1,39 @@
 process MULTIQC {
  label 'process_medium'
 
- conda (params.enable_conda ? "bioconda::multiqc=1.10.1" : null)
+ conda (params.enable_conda ? 'bioconda::multiqc=1.11' : null)
  container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
- 'https://depot.galaxyproject.org/singularity/multiqc:1.10.1--py_0' :
- 'quay.io/biocontainers/multiqc:1.10.1--py_0' }"
+ 'https://depot.galaxyproject.org/singularity/multiqc:1.11--pyhdfd78af_0' :
+ 'quay.io/biocontainers/multiqc:1.11--pyhdfd78af_0' }"
 
  input:
- path 'multiqc_config.yaml'
- path multiqc_custom_config
- path software_versions
+ path ch_multiqc_config
+ path ch_multiqc_custom_config
+ path software_versions_yaml
  path workflow_summary
+ path ('fastqc/*')
  path ("STAR/*")
  path ("salmon_alevin/*")
 
  output:
- path "*multiqc_report.html" , emit: report
- path "*_data" , emit: data
- path "*variants_metrics_mqc.csv", optional:true, emit: csv_variants
- path "*assembly_metrics_mqc.csv", optional:true, emit: csv_assembly
- path "*_plots" , optional:true, emit: plots
+ path "*multiqc_report.html", emit: report
+ path "*_data" , emit: data
+ path "*_plots" , optional:true, emit: plots
+ path "versions.yml" , emit: versions
 
  script:
- def custom_config = params.multiqc_config ? "--config $multiqc_custom_config" : ''
  def args = task.ext.args ?: ''
+ def custom_config = params.multiqc_config ? "--config $multiqc_custom_config" : ''
  """
- multiqc -f $args $custom_config .
+ multiqc \\
+ -f \\
+ $args \\
+ $custom_config \\
+ .
+
+ cat <<-END_VERSIONS > versions.yml
+ "${task.process}":
+ multiqc: \$( multiqc --version | sed -e "s/multiqc, version //g" )
+ END_VERSIONS
  """
 }

nextflow.config

@@ -48,7 +48,8 @@ params {
  igenomes_base = 's3://ngi-igenomes/igenomes'
  igenomes_ignore = false
 
- // MultiQC options
+ // QC and MultiQC options
+ skip_fastqc = false
  multiqc_config = null
  multiqc_title = null
  max_multiqc_email_size = '25.MB'

nextflow_schema.json

@@ -373,6 +373,10 @@
  "hidden": true,
  "fa_icon": "fas fa-bacon"
  },
+ "skip_fastqc": {
+ "type": "boolean",
+ "description": "Skip the FastQC reporting feature in the pipeline."
+ },
  "skip_multiqc": {
  "type": "boolean",
  "description": "Skip the MultiQC reporting feature in the pipeline."

subworkflows/local/fastqc.nf

@@ -0,0 +1,38 @@
+//
+// Check input samplesheet and get read channels
+//
+include { FASTQC } from '../../modules/nf-core/modules/fastqc/main'
+
+workflow FASTQC_CHECK {
+ take:
+ ch_fastq
+
+ main:
+ ch_fastq
+ .map { ch -> [ ch[0], ch[1] ] }
+ .set { ch_fastq }
+
+ /*
+ * FastQ QC using FASTQC
+ */
+ FASTQC ( ch_fastq )
+ fastqc_zip = FASTQC.out.zip
+ fastqc_html = FASTQC.out.html
+
+ fastqc_zip
+ .map { it -> [ it[1] ] }
+ .set { fastqc_zip_only }
+ fastqc_html
+ .map { it -> [ it[1] ] }
+ .set { fastqc_html_only }
+
+ fastqc_multiqc = Channel.empty()
+ fastqc_multiqc = fastqc_multiqc.mix( fastqc_zip_only, fastqc_html_only )
+ fastqc_version = FASTQC.out.versions
+
+ emit:
+ fastqc_zip
+ fastqc_html
+ fastqc_version
+ fastqc_multiqc
+}

workflows/scrnaseq.nf

@@ -36,6 +36,7 @@ ch_multiqc_custom_config = params.multiqc_config ? Channel.fromPath(params.multi
 // SUBWORKFLOW: Consisting of a mix of local and nf-core/modules
 //
 include { INPUT_CHECK } from '../subworkflows/local/input_check'
+include { FASTQC_CHECK } from '../subworkflows/local/fastqc'
 include { KALLISTO_BUSTOOLS } from '../subworkflows/local/kallisto_bustools'
 include { SCRNASEQ_ALEVIN } from '../subworkflows/local/alevin'
 include { STARSOLO } from '../subworkflows/local/starsolo'
@@ -107,6 +108,14 @@ workflow SCRNASEQ {
 
  ch_versions = ch_versions.mix(INPUT_CHECK.out.versions)
 
+ // Run FastQC
+ ch_multiqc_fastqc = Channel.empty()
+ if (!params.skip_fastqc){
+ FASTQC_CHECK ( ch_fastq )
+ ch_versions = ch_versions.mix(FASTQC_CHECK.out.fastqc_version)
+ ch_multiqc_fastqc = FASTQC_CHECK.out.fastqc_multiqc.ifEmpty([])
+ }
+
  // Run kallisto bustools pipeline
  if (params.aligner == "kallisto") {
  KALLISTO_BUSTOOLS(
@@ -181,15 +190,17 @@ workflow SCRNASEQ {
  )
 
  if (!params.skip_multiqc) {
+
  ch_workflow_summary = Channel.value(
  WorkflowScrnaseq.paramsSummaryMultiqc(workflow, summary_params)
  ).collectFile(name: 'workflow_summary_mqc.yaml')
 
  MULTIQC(
  ch_multiqc_config,
- ch_multiqc_custom_config,
+ ch_multiqc_custom_config.collect().ifEmpty([]),
  CUSTOM_DUMPSOFTWAREVERSIONS.out.mqc_yml.collect(),
  ch_workflow_summary,
+ ch_multiqc_fastqc.collect{it[0]}.ifEmpty([]),
  ch_multiqc_alevin,
  ch_multiqc_star
  )