Skip to content

Running & Reporting

Once your data is prepared and you've selected a profile, you're ready to perform the full analysis for your experiment.

1. Execution Patterns

A standard Snakemake invocation for 3t-seq follows a predictable pattern. Using the hpc profile as an example:

pixi run snakemake \
    --profile profiles/slurm \
    --configfile my_config.yaml

The pipeline will then automatically:

  1. Validate your configuration and sample sheets.
  2. Download genomic references (if not already local).
  3. Deploy Conda environments or Singularity containers.
  4. Execute the rules in the optimal order for your cluster.

Overriding Parameters

Sometimes you may want to override a single parameter without changing your config.yaml. You can do this using the --config flag:

pixi run snakemake --profile profiles/slurm --configfile my_config.yaml --config globals:results_folder="results/rerun_test"

2. Detailed Reporting

Snakemake can generate a comprehensive, self-contained HTML report that includes quality metrics, statistics, and interactive plots for every stage of your pipeline.

Generating the Report

To ensure the report accurately reflects the run that generated the results, use the exact same profile and configuration flags in your reporting command:

pixi run snakemake \
    --report report.html \
    --profile profiles/slurm \
    --configfile my_config.yaml

What's in the Report?

The generated report.html is a stand-alone file that you can share with colleagues. It includes: - Statistics: Read counts, mapping percentages, and duplication rates. - QC Plots: FastQC, MultiQC, and library-specific metrics. - Provenance: The exact versions of tools, environments, and configuration parameters used for the run. - Workflow Steps: An interactive diagram of the analysis steps.

3. Monitoring Your Run

You can monitor the progress of your jobs in real-time: - Terminal Output: Snakemake provides a live count of finished/remaining jobs. - Log Files: Every rule execution creates a corresponding log file in your results/logs/ directory.

Next Steps

For a complete reference of every available parameter, visit the detailed configuration guide: