Get Started
scVIVA-Tools is a consolidated spatial transcriptomics analysis toolkit built on top of
scvi-tools, exposing scVIVA, ResolVI, DestVI, DiagVI, Harreman, gimVI,
Stereoscope, and Tangram through one AnnData-first API.
Installation
pip install scviva-tools
Optional extras:
pip install "scviva-tools[spatial]" # SpatialData + squidpy integration
pip install "scviva-tools[rapids]" # GPU-accelerated neighbor graphs (cuML/cuPy/cuGraph)
pip install "scviva-tools[tutorials]" # jupyter, matplotlib, seaborn
pip install "scviva-tools[all]" # everything above
Quick Start
import scviva
# ResolVI: denoising and segmentation-error correction
scviva.ResolVI.setup_anndata(adata, layer="counts")
model = scviva.ResolVI(adata)
model.train()
adata.obsm["X_resolvi"] = model.get_latent_representation()
Swapping in scviva.SCVIVA for niche-aware differential expression, or scviva.DestVI for Visium
deconvolution, follows the same setup → train → extract pattern.
Resources
- Tutorials, user guide, and API reference are available in the full documentation.
- To report bugs or request features, open an issue on GitHub.
- For usage questions, see GitHub Discussions.