SpaRank

Transferable spatial omics deconvolution.

SpaRank turns expression profiles into rank-ordered token sequences and runs them through a Transformer encoder to predict cell-type composition at each spatial spot. Trained once on a multi-context single-cell reference, a single model deploys directly to spatial sections spanning contexts without per-slice retraining. The same pipeline handles unimodal (RNA), multimodal (RNA + ADT, …), and context-conditioned setups.

Getting Started

• Installation
  • Install torch
  • From PyPI
  • From source
• Quick start
  • Unimodal RNA
  • Multimodal RNA + ADT

Tutorials

• Mouse Isocortex Deconvolution with SpaRank
  • 1. Data
  • 2. Training
  • 3. Load pre-trained checkpoint
  • 4. Quantitative evaluation
• Human Lymphoid Tissue Deconvolution with SpaRank
  • 1. Data
  • 2. Training
  • 3. Load pre-trained checkpoint
  • 4. Predict on spatial datasets
• Human Thymus Multimodal Deconvolution
  • Training
  • Evaluation

API Reference

• API Reference

Indices

• Index

• Module Index