Welcome to our fork of RetroTector, a tool for detecting endogenous retroviruses in genomic sequences.
This repository contains our customized version of RetroTector maintained by the Department of Genomic Medicine, Rigshospitalet, Denmark. We have enhanced the original application by creating a Docker container to simplify deployment and usage across various computing environments, including HPC clusters.
Our goal is to ensure the software is robust, reproducible, and easy to integrate into modern bioinformatics pipelines.
-
Containerized Application:
We provide a Docker image that packages RetroTector and all its dependencies, including Java 21, ensuring consistent execution regardless of the host system. -
Parallel Execution:
Our Docker Compose setup supports running multiple instances concurrently with separate mounted workspaces. -
Configuration Automation:
The container entrypoint automatically updates configuration files to reflect the runtime environment, enabling seamless operation without manual tweaks. -
Cluster Ready:
The container is designed to be compatible with HPC environments and can be converted to Singularity images for cluster deployment.
docker build -t retrotector .docker run --rm -it \
-v /path/to/sample1/:/opt/RetroTector/ReTe1.0.1/Workplace/ \
-w /opt/RetroTector/ReTe1.0.1 \
retrotector bashWe provide a docker-compose.yml example to run four instances in parallel, each mounted to a different sample directory.
docker-compose up --buildTo deploy on HPC clusters, convert the Docker image to a Singularity image:
singularity build retrotector.sif docker-daemon://retrotector:latestThe configuration file /opt/RetroTector/ReTe1.0.1/Database/Config.txt is automatically updated in the container to set the correct working directory.
The container runs RetroTector commands:
java -cp RetroTector101.jar retrotector.RetroTectorEngine SweepDNA quit
java -cp RetroTector101.jar retrotector.RetroTectorEngine SweepScripts quitMount your input data to /opt/RetroTector/ReTe1.0.1/Workplace/ inside the container.
For questions or contributions, please contact:
Department of Genomic Medicine
Rigshospitalet, Copenhagen, Denmark
Email: [frederik.otzen.bagger@regionh.dk]