Dragonflye Nextflow Pipeline

This repository contains a Nextflow pipeline to run Dragonflye for genome assembly using Nanopore and Illumina reads. The pipeline is built following the nf-core framework standards and Dockerized for reproducibility and ease of use.

Requirements

To run this pipeline, you need the following installed on your system:

• Nextflow
• Docker
• Conda

Inputs

The pipeline requires the following inputs:

• A folder containing Nanopore reads (e.g., --nanopore_reads_dir /path/to/nanopore_reads)
• A folder containing Illumina reads (e.g., --illumina_reads_dir /path/to/illumina_reads)
• A text file with sample IDs (e.g., --sample_ids /path/to/sample_ids.txt)
• An output folder name (e.g., --output_folder /path/to/output_folder)

Running the Pipeline

1. Clone this repository:

   git clone https://github.com/your-username/dragonflye_nfcore_pipeline.git
   cd dragonflye_nfcore_pipeline

2. Build the Docker image:

   docker build -t dragonflye_nfcore_pipeline .

3. Run the pipeline using Nextflow with Docker:

   nextflow run workflows/main.nf --nanopore_reads_dir /path/to/nanopore_reads --illumina_reads_dir /path/to/illumina_reads --sample_ids /path/to/sample_ids.txt --output_folder /path/to/output_folder -profile docker

Default Parameters

The default parameters used in the pipeline are:

• CPUs: 32
• RAM: 120 GB
• Genome size: 5.5M
• Racon polish steps: 1
• Medaka polish steps: 1
• Medaka model: r941_min_sup_g507
• Polypolish steps: 1
• Pilon steps: 0

Customizing the Pipeline

If you need to change any default parameters, you can do so by editing the nextflow.config file directly or by passing them as command-line arguments when running the pipeline.

Example Command

nextflow run workflows/main.nf --nanopore_reads_dir /path/to/nanopore_reads --illumina_reads_dir /path/to/illumina_reads --sample_ids /path/to/sample_ids.txt --output_folder /path/to/output_folder --cpus 16 --ram 64 --gsize '6M' -profile docker

Credits

This pipeline extends the Dragonflye pipeline by Robert A. Petit III and leverages best practices from the nf-core framework.