Update nfcore modules in spatialaxe#184
Conversation
XENIUM_PATCH_STITCH failed with 'unrecognized arguments: --min-transcripts-per-cell 50'. When stitch_transcripts.py was split into xenium_patch_stitch_transcripts.py + xenium_patch_stitch_postprocess.py, the --min-transcripts-per-cell argument and its small-cell filter were dropped from the transcripts script, but conf/modules.config still injects the flag. - restore the argument + post-stitch small-cell filter (reassign transcripts of sub-threshold cells to noise, drop their polygons) - repoint tests/test_xenium_patch/test_stitch_transcripts.py from the deleted modules/.../resources/usr/bin path to bin/ (broken since the PR nf-core#154 refactor) - add TestMinTranscriptsFilter: filter behaviour, no-op control, and a subprocess CLI test reproducing the exact production invocation - remove the orphan bin/stitch_transcripts.py (superseded, unused) Validated in the module's sopa container: 15 passed.
fix(xenium_patch): restore --min-transcripts-per-cell in stitch script
| - .github/workflows/awsfulltest.yml | ||
| - .github/workflows/awstest.yml |
There was a problem hiding this comment.
this should be added back, so we can test/make sure that it runs in the cloud
There was a problem hiding this comment.
@mashehu , these files are still in the repo, these were just generating warnings while linting runs, should i remove it from the .nf-core.yml?
There was a problem hiding this comment.
yes.
let's see what kind of warnings you get with them in 🙂
| @@ -1,8 +1,10 @@ | |||
| lint: | |||
| actions_ci: false | |||
There was a problem hiding this comment.
this one should also not be skipped
| withName: PROSEG { | ||
| publishDir = [ | ||
| path: "${params.outdir}/${params.mode}/proseg/preset", | ||
| path: "${params.outdir}/${params.mode}/proseg/proseg", |
There was a problem hiding this comment.
Why the double proseg folder? I would find it better to keep then just once proseg/
|
|
||
| // Step 2b: Convert parquet to CSV (Baysor Julia Parquet.jl incompatibility) | ||
| PARQUET_TO_CSV ( ch_patches ) | ||
| PARQUET2CSV ( ch_patches, ".csv" ) |
There was a problem hiding this comment.
I would remove ".csv". The script has as default anyway ".csv" extension.
|
|
||
| // convert parquet to csv | ||
| PARQUET_TO_CSV(ch_transcripts_file, ".csv") | ||
| PARQUET2CSV(ch_transcripts_parquet, ".csv") |
There was a problem hiding this comment.
Same here. Remove the argument ".csv"
heylf
left a comment
There was a problem hiding this comment.
Some minor comments also from me.
template v4.0.2changesBaysor- Bayorrun,segfree,previewProseg-proseg,proseg2baysorOPT-flip,track,statmeta.ymlfilessnapshotcloses #178