RAD-seq

This repository contains a Snakemake pipeline for preprocessing RAD-seq data: quality control, adapter trimming, rRNA filtering, genome mapping, deduplication, and feature counting.

Overview

• Workflow: Implemented in Snakefile.
• Configuration: Provided by config.yaml (edit paths there to match your system).
• Input: Paired-end FASTQ files expected in raw_data/ named like <SAMPLE>_S1_R1_001.fastq.gz and <SAMPLE>_S1_R2_001.fastq.gz.

Required Software The pipeline invokes a number of command-line tools. Install these (preferably via conda/mamba from bioconda and conda-forge) and make them available in your PATH, or set their absolute paths in config.yaml.

• snakemake (to run the workflow)
• hisat-3n and hisat-3n-table (genome/rRNA mapping and table generation)
• samtools (BAM processing, indexing, flagstat)
• umi_tools (deduplication)
• cutadapt (adapter trimming)
• featureCounts (from the subread package)
• multiqc (aggregate QC reports)
• falco (used here for FastQC-like reports; the pipeline expects a falco binary path in config.yaml)
• variant motif (used by the calculate_table_* rules; ensure variant and the motif subcommand are available)

Recommended (example) conda/mamba environment creation:

# using mamba (faster) if available
mamba create -n radseq -c conda-forge -c bioconda \
	snakemake samtools cutadapt umi_tools subread multiqc hisat-3n hisat3n-table \
	python=3.10 -y
conda activate radseq

# Install falco or other QC tools if not available via conda. Adjust config.yaml to point to the binary.

If some packages are unavailable under the given names in your channels, install them from their project pages or place their binaries somewhere on PATH and update config.yaml accordingly.

Configuration

• Edit config.yaml to set the paths to binaries and reference files the workflow needs. Keys used in Snakefile include:
  • falco, umi_tools, cutadapt, featureCounts
  • ref_genome, ref_genome_fa, ref_rRNA, ref_rRNA_fa, chrom_sizes, ref_hg_genome_gtf

Note: Snakefile contains configfile: "workflow/config.yaml" — if your config.yaml is at the repository root, run Snakemake with --configfile config.yaml (example below) or move/copy your config to workflow/config.yaml.

How to run

1. Do a dry-run to check what will run and that inputs exist:

snakemake -n -s Snakefile --configfile config.yaml

2. Run the full workflow using multiple cores (example 16 cores):

snakemake -s Snakefile --configfile config.yaml --cores 16

3. To execute a single target rule (e.g., mapping to genome tables):

snakemake INTERDIR/hisat3n_table/genome/<SAMPLE>.hisat3n_table.tsv -s Snakefile --configfile config.yaml --cores 8

Notes & Troubleshooting

• Ensure FASTQ filenames match the sample names listed in Snakefile or update the SAMPLE list in Snakefile.
• If a binary path is set in config.yaml, Snakemake will use that path (e.g., umi_tools in config). Update these paths when moving the pipeline between systems.
• Logs and intermediate files are written under preprocess/interdir and preprocess/outdir per the Snakefile constants.
• If hisat-3n or hisat-3n-table are not found, either install them and add to PATH, or set their paths in config.yaml and adjust the Snakefile if necessary.

Outputs

• QC reports: report_qc/ (MultiQC reports)
• Counts: preprocess/outdir/feature_counts/counts_hg_genome.txt
• Calculated pileup tables: preprocess/outdir/calculated_table/{genome|rRNA}/

Contact / Changes

• If you want help adapting this pipeline for new samples or a different reference, open an issue or contact the author.