TRACERx NUGEN RRBS DNAm Pipeline

Repository structure:

/bin: Contains executables and "primary" scripts. Eg: removeN6duplicates.py from Nugen
/conf: Nextflow configuration files. These are used to specify settings for local, debug, and HPC environments.
/inventory: contains an example of the inventory used as input for the pipeline
/modules: Nextflow modules. These follow the naming convention .module.nf, and may be nested within a directory with additional required scripts and files.
/subworkflows: Contains nested subworkflows (eg. FASTQ_preprocess, BISMARK_alignment, NUGEN_deduplication or SUMMARY)
/workflows: Contains the DNAm_nf pipeline

Pipeline Overview:

Description:

This pipeline performs the alignment and methylation extraction for RRBS Nugen sequencing.

Input

The initial input of this pipeline is a CSV file containing the inventory of samples to run (each row corresponding to an independent sample). See inventory/inventory.csv for an example.

DNAm (WORKFLOW)

This workflow has the following subworkflows:

FASTQ_preprocess (Subworkflow): Performs Trimming using trimgalore
BISMARK_alignment (Subworkflow): Performs Bismark alignment and sorts
NUGEN_deduplication (Subworkflow): Performs Deduplication step using Nugen Custom py script (See manual: https://www.nugen.com/sites/default/files/M01394_v6_User_Guide%3A_Ovation_RRBS_Methyl-Seq_System_5912.pdf) and cleanup
SUMMARY (Subworkflow): Aggregates QC reports of all previous processes into a single summary report

Output

The pipeline will output a nested directory in the defined output directory with the following structure:

/bam: Contains bams and bam.ai files
/methylation_extraction: Contains output for Bismark Methylation Extraction (each sample in a subfolder)
/qc: Contains reports for each process (deduplication, Bismark alignment, Bismark methylation extraction, and trimming)
- /dedup
- /mapping
- /methylation_extraction
- /trimming

Additionally, it will output an updated inventory with a new column corresponding to the absolute path to the bam file (sample_merged2.txt) and a summary report (summary_report.tsv) containing some metrics used for QC (eg. aligned reads, mapping efficiency, duplication rate, etc).

Usage:

1. Clone the repo

git clone https://github.com/ccastignani/RRBS_DNAm_pipeline

2. Installation and edit config file

Before you start running the pipeline, you should make sure all required software is installed and the config file is set up accordingly (conf/crick.conf). You should make sure the params section contains the paths to the input and output directories as well as the executables (Trimgalore, Bismark, Bowtie and Samtools).

3. Execution wrappers and launch Pipeline

The pipeline can be called from a bash wrapper script as follows which can be launched via sbatch submit.sh:

#!/usr/bin/env bash
ml Nextflow
ml Singularity
NXF_VER=20.12.0-edge nextflow run ./main.nf -profile crick -queue-size 500 -resume

Acknowledgments:

Special thanks to Mark S. Hill for troubleshooting and nf guidelines and Elizabeth Larose Cadieux for the Nugen RRBS pipeline conceptualization

ccastignani/RRBS_DNAm_pipeline