RNASeq-NF is an NGS analysis pipeline for RNA expression quantification and germline variant calling (GATK4).

The pipeline performs the following tasks.

• Read QC (FastQC)
• Sequence trimming (TrimGalore)
• rRNA removal (SortMeRNA)
• Mapping and read-group annotation (STAR)
• Alignment-QC (RSeQC, Preseq)
• PCR duplicate detection (Sambamba MarkDup)
• Gene-expression/biotype quantification ( featureCounts)
• Gene-expression normalization ( edgeR, DESeq2)
• Transcript quantification (Salmon)
• Variant calling (GATK4)
• QC report (MultiQC, CustomQC)

This implementation is a work in progress and aims to reach feature parity with the UMCU RNASeq pipeline while also introducing new features and methods according to developments in the field.

Several components have been adapted from the nf-core rnaseq community pipeline and rewritten in DSL2 syntax to enable a more modular setup.

Please refer to the nf-core rnaseq manual for a description on how to interpret the different output files.

Download the Nextflow binary.

Install Singularity on the host system.

For UMCU users, please follow the instructions on the HPC wiki on how to use Slurm & Singularity.

Start an interactive Slurm session on the HPC cluster.

srun -n 2 --mem 5G --time 12:00:00 --gres=tmpspace:10G --pty bash

The nextflow process needs to remain active until the analysis (see 4) is finished and all jobs have been scheduled. It is therefore wise to execute the above command within a terminal multiplexer, such as screen or Tmux. Alternatively, the command can be embedded within a sbatch script. Though this should be done by default, ensure that the singularity environment variables point to $TMPDIR. In the srun command above, we passed 10G to our session.

SINGULARITY_LOCALCACHEDIR=${TMPDIR}
SINGULARITY_TMPDIR=${TMPDIR}

Clone this repository and ensure that the master branch is checked-out.

git clone --recurse-submodules https://github.com/UMCUGenetics/RNASeq-NF.git

3.1 Resource files
3.2 Settings overview
3.3 Analysis

Before starting the pipeline, ensure that the input fastq files follow the Illumina Naming Convention.

Run the pipeline.

./nextflow run ./RNASeq-NF/main.nf -c </path/to/run.config> --fastq_path <fastq_dir> --out_dir <output_dir> --email <email> -profile slurm -resume 

For local execution (without HPC backend), simply omit the -profile parameter.