Tutorials

Tutorials walk through complete analyses from raw STARsolo output to biological results. They are learning-oriented: every step is shown, every intermediate object is described.

---

Planned for v1.1

Quickstart: end-to-end splicing analysis (Jupyter notebook)

A self-contained notebook that:

1. Downloads a small public STARsolo SJ output (PBMC 3k, ~200 cells after whitelist filtering).
2. Runs read_starsolo → make_m2 → highly_variable_events.
3. Computes logit-PSI, runs PCA via scanpy, and clusters cells.
4. Demonstrates pseudo_correlation against a gene-expression PC1 derived from the same cells.

The notebook will be rendered and embedded here once the v1.1 release ships. Until then, see Getting started for a condensed walkthrough using synthetic data.

---

Notebook execution

All tutorial notebooks are executed at documentation build time using mkdocs-jupyter. Committed notebooks with stale outputs are not accepted; the CI workflow re-runs them from scratch on each merge to main.