Install Planemo

python -m venv planemo
. planemo/bin/activate
pip install planemo

Initialize chap.xml

planemo tool_init --id 'chap' --name 'CHESS Analysis Pipeline' --example_command 'runner.py --config config.yaml' --example_input config.yaml --cite_url 'https://github.com/CHESSComputing/ChessAnalysisPipeline' --help_from_command 'runner.py --help' --test_case --example_output data.out

this command will output:

Tool written to chap.xml
No test-data directory, creating one.
Copying test-file config.yaml

Perform linting:

planemo l
Linting tool /Users/vk/Work/CHESS/ChessPipeline/chap.xml
Applying linter tests... CHECK
.. CHECK: 1 test(s) found.
Applying linter output... CHECK
.. INFO: 1 outputs found.
Applying linter inputs... CHECK
.. INFO: Found 2 input parameters.
Applying linter help... CHECK
.. CHECK: Tool contains help section.
.. CHECK: Help contains valid reStructuredText.
Applying linter general... CHECK
.. CHECK: Tool defines a version [0.1.0+galaxy0].
.. CHECK: Tool defines a name [CHESS Analysis Pipeline].
.. CHECK: Tool defines an id [chap].
.. CHECK: Tool specifies profile version [21.05].
Applying linter command... CHECK
.. INFO: Tool contains a command.
Applying linter citations... CHECK
.. CHECK: Found 1 likely valid citations.
Applying linter tool_xsd... CHECK
.. INFO: File validates against XML schema.

Now, we can start server via the following commad: planemo s, it will take a while. Once finished we may visit http://127.0.0.1:9090 to see our galaxy hub along with our pipeline tool.

Run galaxy

The galaxy framework can be run in two different ways:

  • either via planemo s command, or

  • by cloning galaxy repository and run run.sh script

run galaxy via planemo

planemo provides an easy wasy to run galaxy via planemo s (start the server) command. But when you invoke this command it will fetch all necessary dependencies for galaxy and this process will take 1/2 hour. Unfortunately, this process does not cache dependencies, and it will be repeated upon next call.

Therefore, we recommend to run galaxy natively.

run galaxy natively

To run galaxy natively you must clone its repository:

git clone https://github.com/galaxyproject/galaxy.git

After that, just run run.sh and it will setup python virtual environment in $PWD/.venv, fetch and build all dependencies. This is one time process and after next call the galaxy server will start quickly.

Upon successfull start you will see the following:

Serving on http://127.0.0.1:8080

and can visit this URL to access your galaxy server.

Use conda for galaxy

The galaxy by default relies on python virtual env and will install it in .venv area. But it also supports conda environment. To use it you need:

  • install anaconda

  • source anaconda env

  • start galaexy as run.sh --no-create-venv or run.sh --skip-venv

The one big advantage of using conda is that it will allow packages to install requirements if they are specified in shed xml file.

Adding tool manually to galaxy

  • perform local install of galaxy via

git clone clone https://github.com/galaxyproject/galaxy.git

  • create your new area within tools

cd tools
mkdir chess
cd chess

  • copy your files to your tools area, e.g.

cp chap.xml data.csv img.png ~/Work/CHESS/GIT/galaxy/tools/chess

  • adjust your galaxy config/tool_conf.xml.sample or config/tool_conf.xml file with your new tool info, e.g.

<section name="CHESS" id="chess">
   <tool file="chess/chap.xml" />
</section>

  • start galaxy via run.sh

For more info see this page

References

  1. Installing tools into Galaxy

  2. Adding custom tool to galaxy

  3. how to publish tool to Shed

  4. Galaxy administration

  5. Galaxy support