Add Logistic PCA (LPCA) analysis

[Jeebjean]

Summary

Adds Logistic PCA (LPCA) as a new SPRAS analysis, reducing the binary
edge-by-run matrix built from pathway reconstruction outputs and reporting the
proportion of deviance explained. LPCA is the binary-data counterpart to the
existing PCA analysis.

Changes

• New analysis.lpca config block (include, k, m, cv, transpose) with
  a matching LpcaAnalysis schema and wiring in config.py.
• lpca_analysis Snakemake rule, restricted to algorithms with multiple
  parameter combinations (same guard as the per-algorithm PCA analysis).
• spras/analysis/lpca.py: builds the binary matrix via summarize_networks
  and runs the logisticPCA container through run_container_and_log. m is
  fixed by default; cross-validation (cv: true) and matrix transposition
  (transpose: true, giving runs x edges to mirror the classic PCA analysis)
  are opt-in.
• docker-wrappers/lpca/: pinned rocker/r-base Dockerfile installing
  logisticPCA and its ggplot2 dependencies, the run_lpca.R / run_cv.R
  scripts, and a README.

Testing

Testing this PR locally

The wrapper is included in this PR, so you can build the image locally and run
LPCA without waiting for anything to be published:

docker build -t reedcompbio/lpca:v1 docker-wrappers/lpca/

Docker uses the local image without pulling, so once it is built, enable
analysis.lpca.include: true in your config and run the pipeline as usual.
I built and tested this end to end locally on pathlinker and omicsintegrator1.

Before / at merge

For users who do not build locally, the image needs to be published to
reedcompbio/lpca:v1 on Docker Hub by a maintainer with organization access:

docker build -t reedcompbio/lpca:v1 docker-wrappers/lpca/
docker push reedcompbio/lpca:v1

cc @agitter @ntalluri

[read-the-docs-community]

Documentation build overview

📚 spras | 🛠️ Build #33534855 | 📁 Comparing bb338a8 against latest (1cb9d03)

  🔍 Preview build  

3 files changed

± genindex.html
± fordevs/spras.config.html
± fordevs/spras.html