revise-svc 0.0.3

Spatial transcriptomics benchmarking and annotation toolkit.

REVISE

REVISE (Regenerative Evaluation of VIable Spatial Expression) is a Python toolkit for benchmarking and annotating spatial transcriptomics (ST) data. It bundles standardized benchmarking workflows and application-ready SVC (Spatial transcriptomics Variational Comparison) pipelines so that researchers can compare algorithms, reproduce analyses, and build their own reference pipelines with minimal friction.

Highlights

• Benchmark module: Reproducible evaluation pipelines for simulated or public datasets, enabling method-to-method comparisons.
• Application module: Annotation, reconstruction, and downstream analyses for real ST data with built-in ST/SC preprocessing.
• Utility tools: Ready-to-use helpers such as efficient similarity metrics in revise.tools.distance, simplifying scripting.
• Extensible architecture: Layered BaseSvc / ApplicationSvc classes make it straightforward to plug in new tasks or methods.

Quick Start

git clone https://github.com/wuys13/REVISE.git
cd REVISE
python -m venv .venv && source .venv/bin/activate
pip install -e ".[annotation]"

Minimal Example

import anndata as ad
from revise.application import SpSvc

st = ad.read_h5ad("data/spatial.h5ad")
sc = ad.read_h5ad("data/single_cell_reference.h5ad")
config = ...

svc = SpSvc(st, sc, config=config, logger=None)
svc.annotate()
svc.reconstruct()

Explore the full set of benchmark/application services in revise/application and revise/benchmark. To extend the framework, inherit from the relevant base class and override preprocessing, optimization, or evaluation hooks.

Repository Layout

• revise/application: SVC workflows for real datasets.
• revise/benchmark: SVC variants for benchmarking studies.
• revise/methods: Algorithm implementations and model components.
• revise/tools: Distance metrics, logging helpers, and general utilities.
• conf: Example configurations and experiment parameters.

Contributing

Issues and pull requests are welcome-especially bug reports, documentation improvements, and new method implementations. Install the dev extras via pip install -e ".[dev]" to run ruff and pytest before submitting changes.

License

REVISE is released under the MIT License (see LICENSE).