Repository navigation
Osteosarc 0.6.1
Documentation rewritten in plain language. No library changes.
python -m pip install --upgrade osteosarc==0.6.1-
Plain language. Every page except the corrections page is shorter and easier to read: everyday words instead of internal jargon, each caveat stated once, and lists instead of dense paragraphs. The README, which is also this PyPI page, is included.
-
Where the data comes from. Key concepts now lists every source:
- osteosarc.com;
- its S3 bucket;
- the website's source repository on GitLab, which supplies eight tables the site is built from but doesn't offer as downloads.
Ensembl, NCBI and UCSC are only used as reference genomes and gene annotation.
-
SV catalogue. The page says which of its 637 candidates come from osteosarc.com files (633) and which were added by hand (4).
-
Review fixes. A code review of these docs found 10 statements that didn't match the code, and all are fixed. Among them:
- recipe source identities;
discovercaching;subset_templatesnever fetching mates;- partner-matching rules;
curation --strictchecking the snapshot rather than the live site;--source-revisionneeding a full SHA.
Validation:
- CI passes on Python 3.9–3.13.
- 253 tests pass against the built wheel.
mkdocs build --strictpasses.- The changed pages' examples pass against live osteosarc.com.
PR #41 · Documentation · PyPI