Dockage: mooring of Dockerfiles 🐳

Welcome to Dockage, a collection of Dockerfiles mainly designed for use in bioinformatic workflows.

Purpose

This repository aims to simplify the process tracking all recipes to assemble our docker images. Once built, these containers can be easily converted to Singularity format and deployed on HPC clusters, offering both portability and reproducibility.

Building Docker Images

Each subdirectory contains a Dockerfile for a specific tool or toolbox/pipeline. To build a Docker image locally:

1. Navigate to the specific tool’s directory (e.g., dockage/toolName/).
2. Run the following docker build command:

docker build -t toolName:version -f Dockerfile.version .

This will create a Docker image tagged as toolName:version that you can use locally or push to a container registry. 3. To further convert that image to a singularity container we recommend:

singularity build toolName-version.sif docker-daemon://toolName:version

Repository Structure

dockage/
│
├── tool1/
│   ├── Dockerfile         # Dockerfile for the latest version or master branch of the tool
│   ├── Dockerfile.0.1.2   # Dockerfile for version 0.1.2 of tool1
│   └── additional-files/  # Optional: supplementary scripts or dependencies
│...
│...
├── scripts/               # Useful scripts for automating the build and conversion process
│   ├── build.sh           # Example build script for Docker images
│   └── convert.sh         # Example script for converting Docker images to Singularity
│
└── README.md              # Documentation and usage instructions

Considerations when creating images

🐍 Python-based images

Make sure you install jupyterlab and papermill.

🐘 R-based images

Make sure you install rstudio.

Dockerfile in docker image

Copy the Dockerfile to the container image as the last step. Add this as the last two lines of your Dockerfile:

COPY Dockerfile /docker/
RUN chmod -R 755 /docker

GPU-base image

For best compatiblity with current drivers on the HPC we recommend using:

FROM nvidia/cuda:12.1.1-cudnn8-runtime-ubuntu22.04

⚠️ Do not use conda when creating containers

Contributing

We welcome contributions! If you have improvements to the existing Dockerfiles or want to add support for more bioinformatics tools, please feel free to open a pull request.