Running a pipeline in Codespaces
You can run Codespaces from any nf-core pipeline repository on Github. It also works for the nf-core tools, modules, and website repositories.
Take the nf-core/rnaseq pipeline for example: It can be test-run from its repository on Github by clicking the green button labelled “Code” and then selecting “Create codespace” from the Codespaces tab. Alternatively, visit github.com/codespaces/new/nf-core/rnaseq to have more control over the configuration. Make sure to also select the larger 4 cpu hardware from “Machine Type”.
Once Codespaces has loaded, a container with all required tools is run automatically and the following command can run the nf-core workflow:
nextflow run . \
-profile test,singularity \
--outdir my_resultWe recommend the use of the singularity profile with -profile singularity to successfully run the pipeline.
Although docker is available and containers can be run inside the nf-core devcontainers, running nextflow with the docker profile using -profile docker is currently not yet supported.
This should run the test data through nf-core/rnaseq, using apptainer with your results in the folder: my_result.
This may take some time to complete.
The devcontainer method in Codespaces makes it easy to run and test nf-core pipelines quickly, but it lacks the computational power required to run real-size datasets.
Testing your module with nf-test
If you enter the Codespaces environment for the modules branch, you can run nf-test in order to debug a particular module.
Once you are in the environment, try running an example nf-test for an existing module:
nf-test test --tag <module_name> --profile singularity