search

BIMSBbioinfo / pigx_rnaseq

Bulk RNA-seq Data Processing, Quality Control, and Downstream Analysis Pipeline
GNU General Public License v3.0
20 stars 11 forks source link

suggesting runtime tests #126

Open smoe opened 2 years ago

smoe commented 2 years ago

There should be some prior art on this somewhere on the Internet - https://www.nist.gov/programs-projects/genome-bottle comes to mind. But that is already about semantics. I thought it would possibly help if we ran pigx-rnaseq on very a small example project for a technical validation. Would you be open for that? I am tempted to suggest to avoid anything human (because of the personal data issue) and take something that geneticists have a good understanding for, like worm or drosophila, and a fragment of reads and the genome would suffice.

borauyar commented 2 years ago

I don't know what would be the technical validation for. Is it for testing the pipeline on a real use case for releases? I run it on real datasets every time I make a release. We also tested this on real use case in the pigx paper, showing that we could reproduce the results in the original paper where the data was published.

smoe commented 2 years ago

After preparing a small patch you may want to run the pipeline quickly to learn how this affects the reports. Also, among developers you want to make sure that everyone looks at the same kind of input. My hope is that this also helps with the reporting of errors.

rekado commented 2 years ago

is the existing test infrastructure (run with make check) not enough? Each of these tests runs the pipeline (or parts of it) on test data. For new features it is always good to have matching tests that demonstrate and validate them.

smoe commented 2 years ago

I need a bit of time to cross-check.