Repository navigation
Osteosarc 0.5.0
Makes the command line easier for exploring samples and their files.
python -m pip install --upgrade osteosarc==0.5.0
osteosarc samples --assay scrna-seq --platform ont # Samples with single-cell Nanopore data
osteosarc assets --sample T1_tumor --kind alignment # Their files, with complete keys- Sequencing uses the filter names.
samplesandspecimensshow sequencing as the names the filters take, such asrna-seq; wes; wgs; scrna-seq (ont, pacbio). Before, they showed registry labels likescRNA_ONTthat the filters don't accept. - Filter the overview by sequencing.
samples --assay/--platformlist the samples with that sequencing, anddescribe_samples()takes the same filters. assetsprints a table. It shows complete file keys (whatreadstakes) and a hint when rows are cut off.--jsonprints the full records.- Bad filter values are errors. An assay, platform or tissue that no file uses now fails with a list of the valid values instead of returning nothing. A registry label names the filters to use:
'scRNA_ONT' is a registry label; select assay 'scrna-seq' with platform 'ont'. - Tables: keys are never cut off, cells wrap without splitting hyphenated names, and sizes show in KB, MB or GB.
Validation:
- CI passes on Python 3.9–3.13, including Varcode 7/9 consumer checks.
- 250 tests pass against the built wheel in a clean virtualenv.
- The changed documentation examples pass against live osteosarc.com.