This is a Nextflow workflow which:
- Derives configuration from an SDRF file
- Determines protocol and associated settings
- Runs quantifications appropriate to the protocol
- Aggregates resulting matrices
- If specified, Runs downstream analysis using Scanpy, leveraging the scanpy-scripts package to run individual steps of the Scanpy workflow.
It's actually a workflow-of-workflows comprising:
- scxa-droplet-quantification-workflow
- scxa-smartseq-quantification-workflow
- scxa-aggregation-workflow
- scxa-tertiary-workflow, via scanpy-scripts here
- scxa-bundle-workflow
These are submodules of scxa-workflows.
Workflow dependencies are managed via conda-forge, Bioconda and/or containers. For conda, so you'll need to set that up, see instructions here.
Obviously you'll need Nextflow itself. If you don't have it already you can install via Conda:
conda install nextflow
You may well want to do this within a Conda environment you create for the purpose.
Clone this repo into a directory called 'workflow' under the directory from where you will execute it, such that you have (RUN DIR)/workflow/scxa-control-workflow.
You will also need to clone scxa-workflows, which contains child workflows as submodules, and acts as a central place to store parameters. Clone it under the same directory, so that you have (RUN DIR)/workflow/scxa-workflows.
Current config setup for SLURM.
Routine analysis is triggered (from the above directories) like:
./workflow/scxa-control-workflow/bin/submitControlWorkflow.sh -t scanpy-NFworkflow
This will look for SDRF files in the directory specified by the environment variable SCXA_SDRF_DIR, triggering analyses for any new experiments found there, running quantifications via Nextflow child workflows, and tertiary analysis.
Outputs will be placed in the directory defined as SCXA_RESULTS under 'env' in nextflow.config ('results' by default).