Here is an attempt at testing outputs of nextflow in test.sh.
It runs the pipeline on a 'test' sample and compares the output with the references. It takes about 2 minutes on my machine. It would be better to test on multiple samples but it would be difficult to do locally and also without using some kind of large file storage. Maybe if this is possible with Github LFS or similar we could implement this in Github CI?
Another option is to use smaller simulated data but I'm slightly wary of this because creating representative simulated data is another challenge in itself.
Any suggesions welcome. Happy to keep this discussion open and have this as a separate branch for now.
Here is an attempt at testing outputs of nextflow in test.sh.
It runs the pipeline on a 'test' sample and compares the output with the references. It takes about 2 minutes on my machine. It would be better to test on multiple samples but it would be difficult to do locally and also without using some kind of large file storage. Maybe if this is possible with Github LFS or similar we could implement this in Github CI?
Another option is to use smaller simulated data but I'm slightly wary of this because creating representative simulated data is another challenge in itself.
Any suggesions welcome. Happy to keep this discussion open and have this as a separate branch for now.