Identifying batch-integrated domains from spatial transcriptomics via graph autoencoder with contrastive learning based on cross-modality and data augmentation
Briefings in Bioinformatics · 04.05.2026 · 27 (3)
DOI: 10.1093/bib/bbag248 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 25.05.2026. Verfasst von Yexuan Mao und 10 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Technology, Soochow University, 333 Ganjiang East Road, Gusu District, Suzhou, Jiangsu 215006, China. Hauptthemen laut MEDLINE: Spatial Transcriptomics, Transcriptome. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China, Jiangsu Colleges and Universities QingLan, University-Industry Collaborative Education Program.
Abstract (Originalsprache)
Spatially resolved transcriptomics (SRT) allows for the comprehensive profiling of gene expression while preserving spatial context, advancing the study of tissue architecture. However, existing computational approaches still face key limitations, particularly the insufficient exploitation of histology information and the lack of cross-modal meaningful contrastive strategies for biological analyses. To overcome these challenges, we propose GCAST, a graph contrastive autoencoder framework for spatial transcriptomics that seamlessly integrates multimodal SRT data. GCAST adopts a self-supervised strategy to derive biologically meaningful representations directly from histology images when available. GCAST constructs dual graph views based on data augmentation and introduces a novel contrastive learning designed to leverage histology-weighted and gene-weighted features and improve biological interpretability. In addition, GCAST employs a block-diagonal graph construction to automatically align multiple datasets, achieving batch-effect correction without manual intervention. The framework not only captures spatial gene expression patterns to identify tissue domains but also adapts to datasets with or without histological images and supports the integration of multiple datasets for joint analyses. Overall, GCAST provides a unified and biologically informed framework that has the potential to facilitate deeper analyses of spatial transcriptomics.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Computer Science and Technology, Soochow University, 333 Ganjiang East Road, Gusu District, Suzhou, Jiangsu 215006, China.
- Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing University, 163 Xianlin Avenue, Qixia District, Nanjing, Jiangsu 210023, China.
- Jiangsu Key Laboratory of Drug Discovery and Translational Research for Brain Diseases, Soochow University, 199 Renai Road, Suzhou Industrial Park, Suzhou, Jiangsu 215123, China.
- China Mobile (Suzhou) Software Technology Company Limited, 58 Kunlunshan Road, Huqiu District, Suzhou, Jiangsu 215163, China.
Förderung
- National Natural Science Foundation of China
- Jiangsu Colleges and Universities QingLan
- University-Industry Collaborative Education Program
- Priority Academic Program Development of Jiangsu Higher Education Institutions
- Collaborative Innovation Center of Novel Software Technology and Industrialization
Zitieren
APA
Mao, Y., Quan, L., Chen, X., Wang, S., Zhang, G., Li, S., Jiang, Y., Nie, L., Wu, T., Meng, L., & Lyu, Q. (2026). Identifying batch-integrated domains from spatial transcriptomics via graph autoencoder with contrastive learning based on cross-modality and data augmentation. Briefings in Bioinformatics, 27(3). https://doi.org/10.1093/bib/bbag248
Vancouver
Mao Y, Quan L, Chen X, Wang S, Zhang G, Li S, et al. Identifying batch-integrated domains from spatial transcriptomics via graph autoencoder with contrastive learning based on cross-modality and data augmentation. Briefings in Bioinformatics. 2026;27(3). doi:10.1093/bib/bbag248
BibTeX
@article{Mao2026,
author = {Mao, Yexuan and Quan, Lijun and Chen, Xiangyu and Wang, Siyuan and Zhang, Guozheng and Li, Siqi and Jiang, Yelu and Nie, Liangpeng and Wu, Tingfang and Meng, Lingkun and Lyu, Qiang},
title = {Identifying batch-integrated domains from spatial transcriptomics via graph autoencoder with contrastive learning based on cross-modality and data augmentation},
journal = {Briefings in Bioinformatics},
year = {2026},
volume = {27},
number = {3},
doi = {10.1093/bib/bbag248}
}
EUVIMED