Chapter 6: Building an End-to-End RNA-Seq Pipeline in Nextflow DSL2

Chapter 6: Building an End-to-End RNA-Seq Pipeline in Nextflow DSL2

In this chapter, we assemble a complete, production-grade Nextflow DSL2 bulk RNA-Seq pipeline from raw paired-end reads to a unified MultiQC summary report.

1. Complete main.nf Pipeline Script

nextflow.enable.dsl = 2

// Include modular subworkflows or processes
include { FASTQC       } from './modules/fastqc.nf'
include { FASTP        } from './modules/fastp.nf'
include { STAR_ALIGN   } from './modules/star_align.nf'
include { FEATURECOUNTS} from './modules/featurecounts.nf'
include { MULTIQC      } from './modules/multiqc.nf'

workflow {
    // 1. Validate inputs
    params.input       = 'samplesheet.csv'
    params.genome_dir  = '/ref/star_index'
    params.gtf         = '/ref/genes.gtf'
    params.outdir      = 'results'

    // 2. Parse sample sheet
    ch_samples = Channel.fromPath(params.input)
        | splitCsv(header: true)
        | map { row -> tuple(row.sample, [file(row.fastq_1), file(row.fastq_2)]) }

    // 3. Raw read QC
    FASTQC(ch_samples)

    // 4. Trimming with Fastp
    FASTP(ch_samples)

    // 5. Alignment with STAR
    ch_star_index = Channel.value(file(params.genome_dir))
    STAR_ALIGN(FASTP.out.reads, ch_star_index)

    // 6. Gene Quantification
    ch_gtf = Channel.value(file(params.gtf))
    FEATURECOUNTS(STAR_ALIGN.out.bam, ch_gtf)

    // 7. MultiQC Summary Report
    ch_multiqc_files = FASTQC.out.zip
        .mix(FASTP.out.json)
        .mix(STAR_ALIGN.out.log)
        .mix(FEATURECOUNTS.out.summary)
        .collect()

    MULTIQC(ch_multiqc_files)
}

2. Executing the Pipeline

# Run locally with Docker container profile
nextflow run main.nf \
    --input samplesheet.csv \
    --genome_dir /ref/star_index \
    --gtf /ref/genes.gtf \
    -profile docker \
    -resume