Installation

Nextflow allows to execute pipelines locally, in a cluster with a scheduler (Slurm, PBS, …) or on the cloud (AWS, Azure…).

The dependencies can be installed in a Conda environment, or using a Docker or Singularity container.

Prerequisites

The pipeline requires:

  • A set of dependencies (provided via Conda or Singularity image or Docker container)
  • A Linux system capable of performing de novo assemblies (a single local machine or an HPC cluster)
  • A set of databases (installable with a script provided in the repository, see here)

Installation

:bulb: Ensure you have Nextflow installed otherwise install it first

  1. Clone the repository (and enter it):
# Download a copy of the repository
git clone https://github.com/MattiaPandolfoVR/MetaPhage
# Enter the repository
cd MetaPhage
  1. Download the database (the default location can be db/ inside the repository)
# Install a required module
pip install wget
# Download the databases in the "db" subdirectory
python bin/python/db_manager.py -o ./db

Dependencies

If using Docker or Singularity, the pipeline is capable of retrieving the appropriate image automatically. For local executions or tests, it is possible to generate a single conda environment (see Miniconda).

Miniconda

:bulb: Ensure that Miniconda is already installed in the system.

cd deps
conda env create -n MetaPhage --file env.yaml

To execute the pipeline, remember to activate the environment first:

conda activate MetaPhage

Docker

A Docker image is available from andreatelatin/metaphage:1.0. Nextflow can fetch it before running the pipeline but if you want to download it before:

docker pull andreatelatin/metaphage:1.0

Singularity

A Docker image is available from docker://andreatelatin/metaphage:1.0. Nextflow can fetch it before running the pipeline but if you want to download it before:

wget "https://s3.climb.ac.uk/ifrqmra-metaphage/v1.0/metaphage.simg"