Bpipe is a workflow manager focused on emulating the simplicity of the command line and making your pipelines look as close to the literal commands that run as possible.
Bpipe pipeline are written in Groovy, a scripting version of Java that enables powerful yet high performing DSLs to be easily developed. The definition of the workflow is in the quantify_rna.groovy file.
To run this workflow, you can first install Bpipe using SDKMan and then execute from within the bpipe directory:
bpipe run quantify_rna.groovy ../test_data/*.fq.gz
Note: please make sure to use Bpipe 0.9.11 or higher for running the example.
Please note the default configuration will run the commands on the local computer, and it assumes that salmon and fastqc are available in the PATH. It is easy to configure Bpipe for other environments, and an example is shown in the bpipe.config file which illustrates how to configure it to run using PBS Torque. To run the Torque version you can tell Bpipe to use the alternative environment like so:
bpipe run --env torque quantify_rna.groovy ../test_data/*.fq.gz
If you would like to see how the HTML report that Bpipe makes looks, add the -r option:
bpipe run -r --env torque quantify_rna.groovy ../test_data/*.fq.gz
This example only uses very simple features of Bpipe. Bpipe has many more features which you can explore in the documentation.
Other useful commands to experiment with are:
bpipe log(see logs of a running pipeline)bpipe stop(stop a running pipeline)bpipe history(show history)
Thanks for trying out the Bpipe example!