title: Workflow management: Snakemake

Reference

Update the workflow/Snakefile as per the need and the config file present in workflow/config.yaml.
Command:

```bash

snakemake -s workflow/Snakefile -j 1 --resources gateway=1 --rerun-incomplete --restart-times 3 --latency-wait 60

snakemake -s workflow/Snakefile plot_all -j 1 --resources gateway=1 --rerun-incomplete --restart-times 3 --latency-wait 60

Force run some rule

snakemake -s workflow/Snakefile -j 1 --resources gateway=2 --rerun-incomplete --restart-times 3 --latency-wait 60 --forcerun stage1Compact

snakemake -s workflow/Snakefile --configfile workflow/config.yaml -j 1 --resources gateway=1 --rerun-incomplete --restart-times 3 --latency-wait 60

If the goal is “rerun plots only, do NOT rerun stage1/stage1Compact”, then restrict allowed rules:

snakemake -s workflow/Snakefile plot_all -j 3 --resources gateway=1 --rerun-incomplete --restart-times 3 --latency-wait 60 -R plots snakemake -s workflow/Snakefile plot_all -j 3 --resources gateway=1 --rerun-incomplete --restart-times 3 --latency-wait 60 --allowed-rules plots -R plots ```
Check summary

```bash snakemake -s workflow/Snakefile --summary

snakemake -s workflow/Snakefile --summary | python workflow/summary_table.py ```
Visualize the DAG

```bash

snakemake -s workflow/Snakefile --dag | dot -Tpng > dag.png

snakemake -s workflow/Snakefile --dag | dot -Tpdf > dag.pdf

snakemake --rulegraph | dot -Tpng > rulegraph.png snakemake --rulegraph | dot -Tpdf > rulegraph.pdf

```

For the plotting code, zpt and mass calibration it is reading the input file path from trials.yaml. But, as I am using the snakemake this looks like not a good idea. I should pass it from the command line so that these steps takes the appropriate input informations.