RICO

RICO (Ribosomal DNA Copy number) is a long-read sequencing–based pipeline for estimating ribosomal DNA (rDNA) copy number (CN) using coverage normalization against single-copy genes (SCGs), with DNA methylation analysis.

The pipeline is designed for Oxford Nanopore Technologies (ONT) whole-genome sequencing data and supports both rDNA CN estimation and CpG methylation profiling.

---

Overview

RICO performs the following major steps:

1. Alignment of unaligned ONT BAMs to a custom reference genome containing rDNA repeats
2. Coverage quantification over rDNA loci and curated SCG panel
3. Estimation of rDNA CN by normalizing rDNA coverage against SCG coverage
4. Extraction of CpG methylation calls within rDNA regions

The workflow is implemented in Nextflow (DSL2) and is optimised for high-performance computing (HPC) environments.

---

Assumptions and current scope

• RICO is currently designed and tested only on NCI Australia (Gadi).
• Job submission uses PBSPro.
• Tool paths are specified as absolute paths to ensure reproducibility.
• Reference files must be downloaded separately (see below)

Tools used in the pipeline

RICO was developed and tested with the following versions:

• Nextflow (25.04.6)
• Minimap2 (2.30)
• Samtools (1.23)
• Bedtools (2.31.1)
• Modkit (0.6.1)

Users should install these tools on Gadi and ensure they are accessible via absolute paths or environment modules (see configuration section below).

---

Installation

Clone the repository on Gadi:

git clone https://github.com/comprna/RICO.git
cd RICO

Load Nextflow on Gadi

module load nextflow/25.04.6
nextflow -version #check

Download reference files (required)

Human and mouse reference genomes and annotation files are available on Zenodo:

DOI: https://doi.org/10.5281/zenodo.18500657

Download the archive and extract:

tar -xzf RICO_ref_v1.tar.gz

Place the extracted ref/ directory inside the cloned RICO repository

RICO currently supports

• Human (GRCh38 + rDNAx5)
• Mouse (GRCm39 + rDNAx5)

Configure

Configure NCI project and storage

Edit nextflow.config and update the following fields to match your NCI project:

process {
  executor = 'pbspro'
  project  = 'jd21' # replace with your NCI project
  storage  = 'gdata/jd21+scratch/jd21+gdata/xc17' # replace with your NCI project
}

• project: your NCI project code
• storage: file systems used by the pipeline (where input data and results are stored)

Configure tool paths

You must install these tools yourself (either via NCI modules, Conda, or local compilation) and update their paths in rico.nf:

# Go to the Tools section
params.minimap2 = "/path/to/minimap2"
params.samtools = "/path/to/samtools"
params.bedtools = "/path/to/bedtools"
params.modkit   = "/path/to/modkit"

Ensure the versions installed are compatible with those listed in the Tools used section above.

Inputs

Unaligned BAMs

Input data are provided via a sample sheet (samples.tsv).

Example samples.tsv:

file_path
/g/data/xc17/zaka/nextflow/rDNA-CN-pipeline/unaligned_bams/test.bam 

Replace the file path with your file!

Run the pipeline

Minimum command (default: human, SCG-2)

From the pipeline directory:

nextflow run rico.nf -config nextflow.config --samplesheet samples.tsv

The results are written to the results folder by default.

But you can specify the results directory using --results_dir:

nextflow run rico.nf -config nextflow.config --samplesheet samples.tsv --results_dir /path/to/output

Other SCGs for human samples

For human samples, three curated SCG panels are provided: SCG-1, SCG-2 (default), SCG-3 Specify using --scg, for example:

nextflow run rico.nf -config nextflow.config --samplesheet samples.tsv --scg 3

If not specified, SCG-2 is used.

Mouse samples

nextflow run rico.nf --samplesheet samples.tsv --species mouse

For mouse samples, it uses a single curated SCG panel only. The --scg parameter is ignored when --species mouse is selected.

Outputs

For each input BAM, RICO produces:

1. Alignment outputs

• <sample>_<reference>.bam
• <sample>_<reference>.bam.bai

2. Copy number estimation

• Coverage outputs for rDNA and SCGs
• rDNA CN estimates

3. Methylation outputs

• BedMethyl file
• Per-read base modification output table for rDNA

Citation

Yuen, Leeder, Hannan, Eyras & Hein. Accurate estimation of ribosomal DNA copy number using nanopore long-read sequencing. (Manuscript in preparation)