Metadata-Version: 2.4
Name: stile
Version: 0.0.1
Summary: Python Boilerplate contains all the boilerplate you need to create a Python package.
Author-email: Harsh Sinha <mail.sinha.harsh@gmail.com>
Maintainer-email: Harsh Sinha <mail.sinha.harsh@gmail.com>
License-Expression: Apache-2.0
Project-URL: bugs, https://github.com/sinhaharsh/stile/issues
Project-URL: changelog, https://github.com/sinhaharsh/stile/blob/master/changelog.md
Project-URL: homepage, https://github.com/sinhaharsh/stile
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: typer
Requires-Dist: scanpy
Requires-Dist: scipy
Requires-Dist: numpy
Requires-Dist: pandas
Requires-Dist: matplotlib
Requires-Dist: shapely
Requires-Dist: hdbscan
Provides-Extra: test
Requires-Dist: coverage; extra == "test"
Requires-Dist: pytest; extra == "test"
Requires-Dist: ruff; extra == "test"
Requires-Dist: ty; extra == "test"
Requires-Dist: ipdb; extra == "test"
Dynamic: license-file

# TMA de-arraying for spatial transcriptomics

STile helps segment tissue microarray (TMA) cores from single-cell spatial transcriptomics data and prepare platform-specific outputs for downstream analysis.

## What it does

- Detects tissue cores by clustering cell centroids with HDBSCAN and refining core labels for Vizgen, Xenium, and CosMx outputs.

## Main entry points

- `stile identify_tissue_cores`: Load a `.h5ad`, cluster cells, assign cores, and save core plots.
- `stile prepare_data`: Convert Vizgen/Xenium/CosMx output directories into an `.h5ad` plus `cell_id`, `x_centroid`, `y_centroid` CSV for further processing.

## Quickstart

```bash
# Build AnnData and centroid CSV from a platform run directory
stile prepare_data xenium path/to/xenium/output path/to/save
stile run
```
