# SURF: A Self‐Supervised Deep Learning Method for Reference‐Free Deconvolution in Spatial Transcriptomics

## TL;DR

SURF is a new deep learning tool that improves analysis of spatial transcriptomics data without needing cell-level references.

## Contribution

SURF introduces a self-supervised deep learning method for reference-free deconvolution in spatial transcriptomics.

## Key findings

- SURF outperforms existing reference-free methods and matches or exceeds reference-based approaches when references are unavailable.
- SURF accurately identifies epithelial-to-mesenchymal transition states in tumor regions of human colorectal liver metastasis datasets.
- The method is robust across different resolutions, species, spatial patterns, and tissue states.

## Abstract

Spatial transcriptomics has revolutionized tissue biology by enabling spatially resolved gene expression profiling. Nonetheless, current spot‐level spatial transcriptomic technologies consolidate signals from multiple cells, complicating cellular‐level analysis. Moreover, matched single‐cell references required by reference‐based deconvolution methods are frequently unavailable. To overcome these limitations, we present SURF, a reference‐free deconvolution tool that integrates high‐dimensional gene data analysis with self‐supervised deep learning to effectively model nonlinear gene interactions and leverage spot relationships. Benchmarking on both synthetic and real datasets shows that SURF consistently outperforms existing reference‐free methods and exceeds reference‐based approaches when appropriate references are absent. Applications across datasets with varying resolutions, species, spatial patterns, and tissue states demonstrate SURF's robust capacity to precisely represent tissue microenvironments. Importantly, SURF successfully identifies clinically significant epithelial‐to‐mesenchymal transition states within tumor regions in a dataset of human colorectal liver metastasis, highlighting its utility in uncovering critical biological mechanisms relevant to disease progression.

SURF is a robust reference‐free deconvolution tool that integrates high‐dimensional spatial transcriptomics gene expression analysis with self‐supervised deep learning, enabling effective modeling of non‐linear gene interactions and spot relationships. SURF excels at delineating tissue architectures and significantly enhances high‐resolution analysis of spatial transcriptomics data, thereby supporting advanced biological discovery and innovative therapeutic research.