Skip to content

Feat/pipeline selector landing page - #1402

Open
jaempawi wants to merge 26 commits into
StatFunGen:mainfrom
jaempawi:feat/pipeline-selector-landing-page
Open

Feat/pipeline selector landing page#1402
jaempawi wants to merge 26 commits into
StatFunGen:mainfrom
jaempawi:feat/pipeline-selector-landing-page

Conversation

@jaempawi

Copy link
Copy Markdown
Contributor

Summary

This PR standardizes the documentation of 42 active SoS module notebooks using a consistent structure:

  • Overview
  • Input
  • Output
  • Minimal Working Example
  • Command Interface
  • Workflow implementation

It improves the method narratives, explains when each module should be run, documents command-line flags and generated files, adds representative previews, and provides runnable sos run examples.

Small APA-calling and PCA example files are included where needed by the documentation.

Scope

  • Active module notebooks only
  • Notebooks moved to code/SoS/graveyard/ are excluded
  • Mini-protocols, vignettes, and command-generator notebooks are excluded
  • The local formatting specification is not included in this PR

Validation

  • Module-format audit: 42/42 passed
  • Notebook JSON parsing: 42/42 passed
  • Leaked project-path scan: 0 occurrences

The branch incorporates the latest main and the existing cTWAS test fix. Upstream workflow changes and graveyard moves were preserved.

jaempawi and others added 26 commits August 4, 2026 16:58
Documented sos run examples pointed at personal checkouts (~/codes/xqtl-protocol/, /home/hs3163/GIT/xqtl-protocol/, ~/GIT/xqtl-protocol/), so they could only work for their author. Replaced the prefixes with plain pipeline/ paths. No other path classes touched.
Answers "which pipelines do I need" from a few questions about the user data, then lists the applicable pipelines with inputs, outputs and commands.

High-dimensional Regression is organised by data level (individual vs summary statistics), with fine-mapping and TWAS weights as products of the same run per Gao. RSS TWAS weights and scEEMS Prediction are marked as not yet wired into SoS.
…atFunGen#1393

genotype_formatting.py, bulk_expression_normalization.py and RNA_calling.py were ported to R and deleted, so the README verification snippet failed at the first step.
MRAID_QTL and polyfun are deprecated; no need to rewrite their paths.
Three references read pipeline/pipeline/reference_data.ipynb; pipelines are run as pipeline/<name>.ipynb from the repo root.
build_flat_book.py collects *.ipynb from code/SoS as pages and copies other files there as assets; a file at the repo root is not staged.
Jupyter Book only builds documents listed in _toc.yml (only_build_toc_files: true) and cannot render a raw .html as a page, so the selector is wrapped in a notebook with the markup in a markdown cell.
Removes the conda step: pixi-setup already provides sos, jupyter and a registered sos kernel, and the old advice to conda activate sos pointed users away from the environment holding plink/samtools/tensorqtl.

Fixes the installer command: piped into bash the two prompts consume nothing, so users silently got a minimal install. Download then run instead.

Replaces the Synapse download with tests/fixtures/, which ships in the repo, and links the pipeline selector.
…r-landing-page

# Conflicts:
#	code/SoS/association_scan/TensorQTL/TensorQTL.ipynb
#	code/SoS/pecotmr_integration/twas_ctwas.ipynb
#	website/_toc.yml
…e' into feat/pipeline-selector-landing-page
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants