DMD vs WT Single-End RNA-seq Pipeline (Nextflow DSL2)

This project is a modular Nextflow pipeline for transcriptome-level comparison of WT vs DMD ΔEx51 mouse tibialis anterior muscle RNA-seq data (single-end).

Dataset Accession IDs

Source: NCBI GEO/SRA

• GEO series: GSE156496
  https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE156496
• SRA project: SRP278118
  https://www.ncbi.nlm.nih.gov/sra?term=SRP278118
• Run accessions mapped in this project:
  • WT1 -> SRR12478073
    https://www.ncbi.nlm.nih.gov/sra/SRR12478073
  • WT2 -> SRR12478074
    https://www.ncbi.nlm.nih.gov/sra/SRR12478074
  • DMD1 -> SRR12478076
    https://www.ncbi.nlm.nih.gov/sra/SRR12478076
  • DMD2 -> SRR12478077
    https://www.ncbi.nlm.nih.gov/sra/SRR12478077

Current samplesheet.csv expects:

sample,fastq_1
WT1,data/WT1_R1.fastq.gz
WT2,data/WT2_R1.fastq.gz
DMD1,data/DMD1_R1.fastq.gz
DMD2,data/DMD2_R1.fastq.gz

Pipeline Steps

1. Read sample sheet (single-end FASTQ input)
2. Run FastQC on raw reads
3. Trim reads with fastp
4. Quantify transcripts with Salmon
5. Aggregate QC with MultiQC
6. Build expression matrix from quant.sf files

Requirements

• macOS/Linux
• Java 17+
• Nextflow (DSL2)
• Conda (Miniforge/Anaconda) for -profile conda

Reference Preparation

Download transcriptome FASTA (example: Ensembl GRCm39 cDNA) and build Salmon index:

mkdir -p refs
salmon index \
  -t refs/Mus_musculus.GRCm39.cdna.all.fa.gz \
  -i refs/salmon_index \
  -k 31

Run Command

From project directory:

export JAVA_HOME=/Library/Java/JavaVirtualMachines/temurin-17.jdk/Contents/Home
export PATH="$JAVA_HOME/bin:$HOME/miniforge3/bin:$PATH"

nextflow run main.nf \
  -profile conda \
  --samplesheet samplesheet.csv \
  --outdir results \
  --salmon_index "$PWD/refs/salmon_index" \
  -resume

Output

• results/fastqc/: FastQC reports
• results/fastp/: trimmed FASTQ + fastp reports
• results/salmon/: sample-level *.quant.sf
• results/multiqc/: MultiQC report
• results/matrix/expression_matrix.tsv: merged expression table

GitHub Repository Note

Large raw data and generated artifacts are intentionally excluded from GitHub, including:

• raw FASTQ files
• Salmon index files
• Nextflow work/ directory
• large intermediate result directories