Alpine scripts - #53
AbhirupaGhosh wants to merge 15 commits into
Conversation
Adding a very simple dev script that creates .rda files out of our .tsv mapping data. Replaced a few places where the only raw TSV data was used instead of the internal datasets. Removed some S. epi files of uncertain origin. Not sure where those came from, honestly.
Updating scripts to use BiocFileCache throughout. Updating manifest for better handoff to amRml. Updating HMMER databases to use BiocFileCache and be saved as resources.
Wee `devtools::document(".")` updates, and adding BiocFileCache to Imports.
Error in `.runHmmerJob()`: ! unused argument (total_proteins = total_proteins) So if make the change, the error is not showing up. But I still don't know the logic.
Without this, clean_drug/drug_class/drug_abbr/class_abbr/cleaned_bvbrc_countries only resolve under devtools::load_all(); a real installed-package library() call throws object 'clean_drug' not found, which breaks runDataProcessing().
Good catch Emily! Can replicate that `checkDataAvailability()` fails with this absent. Co-authored-by: Emily Boyer <130874527+eboyer221@users.noreply.github.com>
CPU resource detection was faulty on HPC. Added parallelly as an Import, and updated to use the recommended future CPU detection method, parallelly::availableCores(). This should behave better to avoid oversubscription on any system, and generally handle CPU requests more carefully.
|
This is a PR but we don't want to merge it. This branch will only be useful for our internal use. I have merged commits from @epbrenner 's current PR and tested its run. |
|
The general recommended steps in Alpine
I have added the Docker images and HMMER databases in |
3 participants
Changed docker to apptainer.
First time run might take longer time since the apptainer will pull the docker images