Running

The pipeline offers a workflow that processes files, images, and builds a Vitessce config file from the generated files. Alternatively, the workflow to process files only, or the workflow to process images only can be called independently.

Each of these workflows work as entry points that can be specified when running the pipeline through the command line.

• The Full_pipeline workflow runs the other workflows as needed and builds a Vitessce config file per dataset.

• The Process_files workflow handles data files and their conversions.

• The Process_images workflow handles image files and/or label image data and their conversions.

The pipeline configuration and the location of input data is specified through the parameters yaml file. This file is passed to Nextflow when running the pipeline.

To run the Full_pipeline use

nextflow run main.nf -params-file /path/to/params.yaml -entry Full_pipeline

This will handle all input files, whether they are data files or images, for all datasets defined.

You can modify the entry point if you’re interested in only getting the converted outputs. Use -entry Process_files or -entry Process_images as you need.

Running using Conda

The default pipeline will run on local executor without any type of environment creation. To run the pipeline using conda environments use the -profile conda option:

nextflow run main.nf \
         -params-file /path/to/params.yaml \
         -entry Full_pipeline \
         -profile conda

Creating the environment when the pipeline is launched may take a few minutes.

Running using Docker

The default pipeline will run on local executor without any type of environment creation. To run the pipeline using Docker containers use the -profile docker option:

nextflow run main.nf \
         -params-file /path/to/params.yaml \
         -entry Full_pipeline \
         -profile docker

Pulling the containers when the pipline is launched may take a few minutes.

Running using Singularity

The default pipeline will run on local executor without any type of environment creation. To run the pipeline using Singularity containers use the -profile singularity option:

nextflow run main.nf \
         -params-file /path/to/params.yaml \
         -entry Full_pipeline \
         -profile singularity

Pulling the containers when the pipline is launched may take a few minutes.

Further reading

For more information about Docker image pulling/local conda env creation in Nextflow please refer to Nextflow’s official docs for containers and conda.