Computational environment to work with Isatools
Contribution to this repository is much appreciated. In order to keep the repository clear and structured, we have a few guidelines:
- If you want to add a feature or see coding parts that can be improved, create a new branch according to the naming conventions
- Use descriptive commit names
This repository is made to demonstrate how to create a virtual environment (conda) and container (Docker) that provides all necessary software to work with ISA-tab files. A template is present on how to work create/modify ISA-tab files.
Directory containg all files needed to build to Docker image. The Dockerfile must be used to build the Docker container. It defines the container's content and structure. The isa_env.yml file is needed to build the conda environment inside the container. The isa_parse.ipynb file is a Jupyter notebook file that demonsrates how to parse ISA-Tab files with isatools.
This directory contains files that describe all steps that have been done prepatory for this repository.vThe create_isa_env_yml.sh file was used to create the conda environment (isa_env) and export this environment to a .yml file, which is used to build the Docker container. The Push_docker_registry.sh file describes how the casperdevisser/isatools_environment image was pushed to the registry.
Directory containing .png files of flowcharts that give a schematic overview of the Docker container
Directory containing the components to build the Singularity container (which can be used in the exact same way as the Docker container). The Dockerfile must be used to build the Docker container. It defines the container's content and structure. The isa_env.yml file is needed to build the conda environment inside the container. The isa_parse.ipynb file is a Jupyter notebook file that demonsrates how to parse ISA-Tab files with isatools. The isa.def file is the Singularity definition file, which must be used to build the singularity container.
cd Docker/
Docker build -t casperdevisser/isatools_environment .
docker pull casperdevisser/isatools_environment:latest
docker run -it -v $(volume):/isa_container/volume -p 8888:8888 casperdevisser/isatools_environment
-
$(volume) is the absolute path to directory where you store input data. Example input:
~/ngms-isa-casperdevisser/Example_input/MTBLS1437_compressed_files. Mount current directory, Windows Command Line: 'docker run -it -v %cd%:/isa_container/volume ...', PowerShell: 'docker run -it -v${PWD}:/isa_container/volume ...', Linux: 'docker run -it -v $ (pwd):/isa_container/volume ...' -
Open
localhost:8888in a browser (outside of the docker). -
If running the container inside the Windows DRE, use the ip-adress that is returned by the command
docker-machine ip. For example:xxx.xxx.xx.xx:8888. -
If token is required for login, copy paste the token that can be found in one of the URLs that is given as output after the docker run command.
Now you can use isatools
Run container in background and print CONTAINER_ID
docker run -it -d isatools_environment
Run image iteractively (use the CONTAINER_ID that was printed with the previous command)
docker exec -it CONTAINER_ID bash
Once inside the container, install new packages to the conda environment:
conda activate isa_env
conda install numpy
Exit the container:
exit
Commit the changes on your local PC terminal and give image new tag (currently, we use the date at the moment of pushing as tag name):
docker commit -m "<describe changes made to container>" CONTAINER_ID casperdevisser/isatools_environment:<new_tag_name> #for example: 12_08_21
Build from isa.def:
cd Singularity
sudo singularity build isa.sif isa.def
Run container:
sudo singularity run --bind /home/<user.name>@mydre.org/isa_action/notebooks:/isa_workdir isa.sif
Opening the lnk to the Jupyter notebook requires the exact same steps as with the Docker image (see 'Run Docker Container' above)