Running the snakemake pipeline

After landmarks have been identified, either using the MAGPIE interactive shiny app or using another method, they must be saved in the structure specified in the Input structure section. Specifically, each set of landmarks must be saved in a comma-separated variable (csv) format, with column names X_left, Y_left, X_right, Y_right where left and right refer to the first and second image respectively.

Once all inputs are in the correct format, you can run the pipeline by navigating to the folder where your Snakemake file, as well as all the input files and scripts, are saved and using:

snakemake

If you want to parallelise the pipeline you can use:

snakemake -j [number of cores]

Other command-line parameters can be found in the snakemake documentation.

Running the pipeline will prompt an input checking stage which ensures that all required inputs are present and of the required structure. A summary of the input formats will be saved in output/summary.csv for reference. The same approach can be used outside the pipeline, for example before using the shiny app to check everything looks okay. For details on subsequent individual pipeline steps, refer to the next documentation pages.