Chapter 8 Illumina QC example
This chapter contains an example nextflow wrokflow for a simple QC of paired Illumina data.
In this example ./ represents the main project/workflow directory.
It will use a locally installed conda env.
8.1 main.nf
The main nextflow file.
Path: ./main.nf
#!/usr/bin/env nextflow
/*
* Workflow for basic Paired end Illumina QC
*/
/ Params in nextflow.config
/Text that will be printed at start of work flow to terminal
log.info """\
ILLUMINA QC
===========
raw reads: ${params.reads}
out_dir: ${params.outdir}
"""
.stripIndent()
/*
* Processes
* All within module files
*/
include { FASTQC_RAW } from './modules/local/fastqc/raw/main.nf'
include { MULTIQC_RAW } from './modules/local/multiqc/raw/main.nf'
include { TRIMMOMATIC } from './modules/local/trimmomatic/main.nf'
include { FASTQC_TRIM } from './modules/local/fastqc/trim/main.nf'
include { MULTIQC_TRIM } from './modules/local/multiqc/trim/main.nf'
/*
* Workflow
*/
/// Read in fastq data into channel
Channel
.fromFilePairs(params.reads, checkIfExists: true)
.set {read_pairs_ch}
/// raw fastqc
FASTQC_RAW(read_pairs_ch)
/// raw fastqc zip output
raw_fastqc_zip_ch=FASTQC_RAW.out.zip
/// Transform so channel of 1 value which is a tuple
/// all R1 in one list and all R2 in one list
raw_fastqc_zip_split_by_r1r2_ch = raw_fastqc_zip_ch.collect(flat:false)
.map{it.transpose()}
/// raw multiqc of fastqc
MULTIQC_RAW(raw_fastqc_zip_split_by_r1r2_ch)
/// trimmomatic
TRIMMOMATIC(read_pairs_ch)
/// trimmomatic fastqc
FASTQC_TRIM(TRIMMOMATIC.out)
/// trim fastqc zip output
trim_fastqc_zip_ch=FASTQC_RAW.out.zip
/// Transform so channel of 1 value which is a tuple
/// all R1 in one list and all R2 in one list
trim_fastqc_zip_split_by_r1r2_ch = trim_fastqc_zip_ch.collect(flat:false)
.map{it.transpose()}
/// trimmomatic_multiqc
MULTIQC_TRIM(trim_fastqc_zip_split_by_r1r2_ch)8.3 Modules
The processes are each in their own separate files/modules.
8.3.1 FASTQC_RAW
Path: ./modules/local/fastqc/raw/main.nf
Note: how in this case the publishDir directive is used to create the output directory where the fastqc files will be located.
#!/usr/bin/env nextflow
process FASTQC_RAW {
conda params.conda_env
publishDir {
"${params.outdir}/raw_fastqc_output"
}, mode:'copy'
input:
tuple val(sample_id), path(reads)
output:
tuple path("./${sample_id}_R1_fastqc.html"),
path("./${sample_id}_R2_fastqc.html"), emit: html
tuple path("./${sample_id}_R1_fastqc.zip"),
path("./${sample_id}_R2_fastqc.zip"), emit: zip
script:
"""
fastqc -o ./ ${reads[0]} ${reads[1]}
"""
}8.3.2 MULTIQC_RAW
Path: ./modules/local/multiqc/raw/main.nf
Output is simple as we don't need to use it in the workflow as it is part of the final output.
#!/usr/bin/env nextflow
process MULTIQC_RAW {
conda params.conda_env
publishDir {
"${params.outdir}/raw_multiqc_output"
}, mode:'copy'
input:
tuple path(r1_zip), path(r2_zip)
output:
path "./r{1,2}/multiqc_report.html"
script:
"""
#R1
mkdir r1
multiqc -o r1 $r1_zip
#R2
mkdir r2
multiqc -o r2 $r2_zip
"""
}8.3.3 TRIMMOMATIC
Path: ./modules/local/trimmomatic/main.nf
Trimmomatic creates:
- P files: Paired reads that retained both pairs.
- U files: Reads that are unpaired as their paired read was removed by the filtering. These are generally not used for most processes.
#!/usr/bin/env nextflow
process TRIMMOMATIC {
conda params.conda_env
publishDir params.outdir, mode:'copy'
input:
tuple path(sample_id), path(reads)
output:
tuple val(sample_id), path("${sample_id}_{1,2}P.fastq"), path("${sample_id}_{1,2}U.fastq")
script:
"""
trimmomatic PE -phred33 \
${reads[0]} ${reads[1]} \
-baseout ${sample_id}.fastq \
SLIDINGWINDOW:4:20 MINLEN:50
"""
}8.3.4 FASTQC_TRIM
Path: ./modules/local/fastqc/trim/main.nf
Only using fastqc on the P files as we are not interested in the U files.
#!/usr/bin/env nextflow
process FASTQC_TRIM {
conda params.conda_env
publishDir {
"${params.outdir}/trim_fastqc_output"
}, mode:'copy'
input:
tuple val(sample_id), path(trim_p_reads), path(trim_u_reads)
output:
tuple path("./${sample_id}_1P_fastqc.html"),
path("./${sample_id}_2P_fastqc.html"), emit: html
tuple path("./${sample_id}_2P_fastqc.zip"),
path("./${sample_id}_2P_fastqc.zip"), emit: html
script:
"""
fastqc -o ./ ${trim_p_reads[0]} ${trim_p_reads[1]}
"""
}8.3.5 MULTIQC_TRIM
Path: ./modules/local/multiqc/trim/main.nf
#!/usr/bin/env nextflow
process MULTIQC_TRIM {
conda params.conda_env
publishDir {
"${params.outdir}/trim_multiqc_output"
}, mode:'copy'
input:
tuple path(r1_zip), path(r2_zip)
output:
path "./r{1,2}/multiqc_report.html"
script:
"""
#R1
mkdir r1
multiqc -o r1 $r1_zip
#R2
mkdir r2
multiqc -o r2 $r2_zip
"""
}