# SCALE: unsupervised multiscale domain identification in spatial omics data

## TL;DR

SCALE is a new method for identifying functional domains at multiple spatial scales in tissue data, improving understanding of tissue organization.

## Contribution

SCALE introduces an unsupervised algorithm combining deep learning and entropy-based search for multiscale domain identification.

## Key findings

- SCALE outperforms existing methods by up to 191.1 percentage points in domain identification.
- It successfully identifies multiscale domains in murine brain and kidney tissue data.
- The method is robust and scalable across different tissue types and platforms.

## Abstract

Single-cell spatial transcriptomics enables precise mapping of cellular states and functional domains within their native tissue environment. These functional domains often exist at multiple spatial scales, with larger domains encompassing smaller ones, reflecting the hierarchical organization of biological systems. However, the identification of these functional domain hierarchies has been largely unexplored due to the lack of suitable computational methods. In this work, we present SCALE, an unsupervised algorithm for multiscale domain identification in spatial transcriptomics data. SCALE combines deep learning-based graph representation learning with an entropy-based search algorithm to detect functional domains at different scales. We demonstrate its effectiveness in identifying multiscale domains using both simulated data and spatial transcriptomics data from murine brain (Xenium and MERFISH) and patient-derived kidney tissue, highlighting its robustness and scalability across diverse tissue types and platforms. SCALE outperforms state-of-the-art multidomain identification by up to 191.1 percentage points. SCALE’s ease of use makes it a powerful aid for advancing our understanding of tissue organization and function in health and disease.

Graphical Abstract