Histology-informed spatial domain identification through multi-view graph convolutional networks

Huihui Zhang, Jiaxing Chang, Zirong Li, Yue Sun, Pinli Hu, Haoxiu Wang, Hang Yang, Yonglin Ren, Xingtan Zhang, Zehua Chen, Kok Wai Wong and Haojing Shao

PLOS Computational Biology, 2026, vol. 22, issue 6, 1-19

Abstract: Identifying spatial domains is crucial in spatial transcriptomics, yet effectively integrating gene expression, spatial location, and histology remains challenging. We present STESH, a Spatial Transcriptomics clustering method that combines Expression, Spatial information and Histology. STESH extracts histological features using a convolutional neural network and generates expression, histology, spatial, and collaborative convolution modules for a multi-view graph convolutional network with a decoder and attention mechanism. We evaluated STESH on multiple tissue types and technology platforms. STESH consistently outperformed ten state-of-the-art methods, achieving superior clustering accuracy with the highest scores in adjusted Rand index, normalized mutual information, and Fowlkes-Mallows index.Author summary: Identifying spatial domains in spatial transcriptomics is key to understanding tissue structure and gene expression patterns, yet existing approaches struggle to fully and effectively combine gene expression data, spatial location information, and histological images—three critical pieces of spatial transcriptomics data. To address this gap, we developed a new method that integrates all three types of information to accurately detect spatial domains. We used deep learning to extract detailed features from histological images, built separate analytical frameworks for each data type, and fused these frameworks with an attention mechanism to capture the intrinsic links between different data modalities. We tested this method on multiple tissue types and technical platforms, comparing it against ten leading approaches, and found it consistently achieved higher clustering accuracy and could precisely reconstruct complex biological tissue structures, even uncovering fine details of tumor heterogeneity. This method enhances downstream spatial transcriptomics analyses and provides a more reliable tool for researchers, helping to deepen our understanding of the spatial organization of tissues and the relationship between gene expression and histological morphology.

Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014281 (text/html)
https://journals.plos.org/ploscompbiol/article/fil ... 14281&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pcbi00:1014281

DOI: 10.1371/journal.pcbi.1014281

Access Statistics for this article

More articles in PLOS Computational Biology from Public Library of Science
Bibliographic data for series maintained by ploscompbiol ().