Running on the Google Cloud Platform

Prepare the pipeline

• activate the snakemake conda environment created in the installation step

  conda activate snakemake

• clone the repository

• modify config/samples.txt and config/config.yaml as described in the inputs and outputs section

• go to workflow directory

  cd <cloned repo dir>/workflow

• Follow the instruction here to create a service account to run the workflow via Google Cloud Life Sciences and then save json credentials and use for the pipeline. For more details, read the snakemake guide as well.

  export GOOGLE_APPLICATION_CREDENTIALS=/path/to/snakemake-credentials.json

• run the pipeline

  snakemake --google-lifesciences --default-remote-prefix <gs_path_for_output> --use-conda --google-lifesciences-region us-east1

• monitor the jobs on gcp console (lifesciences and compute engine)