DiST: disentangling structure via self-supervised tensors for heterophilous spatial transcriptomics.
- 1School of Computer Science, Guangdong University of Technology, No. 100 Waihuan Xi Road, Guangzhou Higher Education Mega Center, Panyu District, 510006, Guangdong, China.
- 2Shenzhen TENCLASS Technology Co., Ltd., 8th Floor, Building 5D, Shenzhen Software Base, No. 11, 13, 15 Haitian 1st Road, Yuehai Subdistrict, Nanshan District, 518000, Guangdong, China.
Abstract
Spatially resolved transcriptomics (SRT) serves as a transformative technology for examining gene expression within tissue architecture. However, accurate spatial domain identification remains challenging due to the inherent data sparsity and the tendency of graph-based methods to over-smooth representations at tissue boundaries. To overcome these limitations, we propose DiST, a novel self-supervised tensor learning framework. Distinct from conventional approaches that rely on neighborhood propagation, DiST introduces a Semantically Aligned Ego-Network Tensor construction strategy. By stacking semantically similar ego-networks from the global context, we transform flat graph data into high-order tensor structures that preserve local heterogeneity. Furthermore, we employ a robust low-rank and sparse decomposition mechanism to separate a coherent component used for representation learning from sparse entrywise deviations associated with sequencing dropouts and technical artifacts. This decomposition yields sparsity-resilient representations that recover sharp tissue structures. Extensive experiments on five benchmark datasets demonstrate that DiST outperforms representative methods in spatial clustering. Moreover, downstream biological validation on human breast cancer data confirms DiST's capability to uncover biologically meaningful molecular signatures and functional heterogeneity.
Related
- Supervised Protein Structure Classification Using Topological Persistence With DeltaFold
- Looking Beyond KD for Aptamer Validation: Disentangling Specific Aptamer Binding from Electrostatic Association for Basic Protein Targets
- Disentangling polymer confinement from specific-folding interactions reveals the drivers of Escherichia coli chromosome organization
- Disentangling Between-Target and Within-Target Signal in PPI Benchmarking
- Disentangling the sucrose metabolism of the corn smut Ustilago maydis reveals unexpected complexity
BibTeX
@article{liu2026dist,
title = {{DiST: disentangling structure via self-supervised tensors for heterophilous spatial transcriptomics.}},
author = {Yaozhang Liu and Yu Hu and Jinyu Liu and Shuaishuai Tan and Xiaoping Li and Guan Huang and Mai Chen and Langsheng Lei and Jun Du},
journal = {Briefings in bioinformatics},
year = {2026},
month = sep,
doi = {10.1093/bib/bbag600},
url = {https://europepmc.org/article/MED/42960993},
note = {CC BY},
}
RIS
TY - JOUR
TI - DiST: disentangling structure via self-supervised tensors for heterophilous spatial transcriptomics.
AU - Yaozhang Liu
AU - Yu Hu
AU - Jinyu Liu
AU - Shuaishuai Tan
AU - Xiaoping Li
AU - Guan Huang
AU - Mai Chen
AU - Langsheng Lei
AU - Jun Du
PY - 2026
DA - 2026/09/01
JO - Briefings in bioinformatics
DO - 10.1093/bib/bbag600
UR - https://europepmc.org/article/MED/42960993
AB - Spatially resolved transcriptomics (SRT) serves as a transformative technology for examining gene expression within tissue architecture. However, accurate spatial domain identification remains challenging due to the inherent data sparsity and the tendency of graph-based methods to over-smooth representations at tissue boundaries. To overcome these limitations, we propose DiST, a novel self-supervised tensor learning framework. Distinct from conventional approaches that rely on neighborhood propagation, DiST introduces a Semantically Aligned Ego-Network Tensor construction strategy. By stacking semantically similar ego-networks from the global context, we transform flat graph data into high-order tensor structures that preserve local heterogeneity. Furthermore, we employ a robust low-rank and sparse decomposition mechanism to separate a coherent component used for representation learning from sparse entrywise deviations associated with sequencing dropouts and technical artifacts. This decomposition yields sparsity-resilient representations that recover sharp tissue structures. Extensive experiments on five benchmark datasets demonstrate that DiST outperforms representative methods in spatial clustering. Moreover, downstream biological validation on human breast cancer data confirms DiST's capability to uncover biologically meaningful molecular signatures and functional heterogeneity.
SN - 1467-5463
SN - 1477-4054
ER -