The COMBAT-TB Workbench is an IRIDA based, module workbench for M. tuberculosis bioinformatics. It is designed to be easily deployed on a single server.
This setup was tested on a VM with the following specs.
- 32G RAM
- 4 vCPU
- 250G Disk
Assumption :
- You have
docker
anddocker-compose
installed on destination instance/VM.
You can use scripts to install docker and deploy this stack.
Connect to the server where you want to install the Workbench:
git clone https://github.com/COMBAT-TB/irida-galaxy-deploy.git ; cd irida-galaxy-deploy
The 'new style' irida-plugin-builder bundles a tools.yaml
with each pipeline in the pipeline jar file. The TB Sample Report and TB Phylogeny pipelines are built using this builder and for each release a pipeline is
published on the corresponding Github repositories.
This repository includes a script update_plugins_and_tools.py
(a Python3 script) that can download these plugins and configure the collection of tools to be installed into Galaxy. This tool takes as input a list of workflows to download (see workflows.txt
) and optionally some extra Galaxy tools not mentioned in the workflows (see extra-galaxy-tools.yml
). Here is a typical run of this tool:
./update_plugins_and_tools.py --extra_tools_file extra-galaxy-tools.yml workflows.txt
This should be run before the docker-compose up
command and whenever workflows are updated or new ones added. It will require rebuilding the Docker containers i.e. docker-compose up --build -d
. The update_plugins_and_tools.py
script also has an option --remove_old_workflows
that will delete all workflows in the workflow directory before downloading new ones. This should be used when new versions of workflows are downloaded.
For the first time you start the Workbench
docker-compose up --build -d
To shut down the Workbench (again from the same directory where the Workbench code is installed):
docker-compose down
To start the Workbench again:
docker-compose up -d
This will take a couple of minutes.. ⌚ ☕
Upon completion, point your browser to:
- REMOTE.SERVER:8080/irida/ to access IRIDA
- REMOTE.SERVER:9090 to access Galaxy
The default administrator username and password are:
admin:password1
for IRIDAadmin:admin
for Galaxy
Please see openstack-terraform to deploy to OpenStack using Terraform.