Snakemake example workflow to test panoptes.

This workflow targets Snakemake 9. Snakemake 9 removed the legacy --wms-monitor flag, so monitoring is now wired up through the snakemake-logger-plugin-panoptes logger plugin.

Runs a small four-sample reference pipeline against the included chr14 GTF and BAM data:

All 14 jobs (4 samples × 3 per-sample rules + 1 merge + 1 finish) and the overall workflow status are streamed to panoptes in real time via --logger panoptes.

The workflow relies on conda dependencies to run, so we recommend installing miniconda first. Instructions: https://docs.conda.io/en/latest/miniconda.html

Install Snakemake 9 and the panoptes logger plugin. With conda (recommended, since the example workflow rules use conda envs anyway):

conda create -n panoptes-example -c conda-forge -c bioconda 'snakemake>=9' snakemake-logger-plugin-panoptes
conda activate panoptes-example

Or with pip in a virtual environment:

python3 -m venv venv
source venv/bin/activate
pip install 'snakemake>=9' snakemake-logger-plugin-panoptes

Then clone this example workflow:

git clone https://github.com/panoptes-organization/snakemake_example_workflow.git
cd snakemake_example_workflow

Start a panoptes server (see the panoptes README) so it is listening on http://127.0.0.1:5000, then:

Create the DAG:

bash create_dag.sh

Run the workflow:

bash run_local.sh

The workflow events are forwarded to panoptes by the --logger panoptes plugin specified in run_local.sh. To point at a panoptes server on a different host or port, edit the --logger-panoptes-address value in run_local.sh (or override it via the SNAKEMAKE_LOGGER_PANOPTES_ADDRESS environment variable).

See the snakemake-logger-plugin-panoptes README for the full list of plugin settings and how it maps Snakemake events to panoptes' API.