EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Biased multi-view contrastive learning with attentive masking for spatial transcriptomic analysis

Laiyi Fu iD, Wenkai Cui, Yifan Chen, Danyang Wu iD, Hequan Sun iD

Briefings in Bioinformatics · 04.05.2026 · 27 (3)

DOI: 10.1093/bib/bbag215 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 11.05.2026. Verfasst von Laiyi Fu und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Automation Science and Engineering, Xi'an Jiaotong University, No. 28 Xianning West Road, Beilin District, Xi'an, Shannxi, 710049, China. Hauptthemen laut MEDLINE: Spatial Transcriptomics, Gene Expression Profiling, Machine Learning, Transcriptome. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Basic Research Plan in Shaanxi Province of China, China Postdoctoral Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Spatial transcriptomics (ST) enables the simultaneous measurement of gene expression and spatial context, offering unprecedented insights into tissue architecture and cellular communication. However, existing approaches often fail to jointly capture spatial topology and transcriptional heterogeneity, leading to suboptimal representations and limited biological interpretability. To address this limitation, we propose stCAMBL, a biased multi-view contrastive framework that integrates spatial graph structure modeling with attentive feature masking and partial contrastive regularization. Built upon a variational graph autoencoder backbone, stCAMBL learns biologically informed and noise-robust embeddings by adaptively emphasizing informative molecular features while mitigating confounding patterns across spatial domains. Comprehensive evaluations on multiple 10$\times$ Visium datasets demonstrate that stCAMBL substantially improves clustering accuracy, gene ontology enrichment, and signal restoration, demonstrating strong generalizability for high-fidelity ST analysis.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Spatial TranscriptomicsGene Expression ProfilingMachine LearningTranscriptomeAutoencoderHumansAlgorithmsClustering Algorithms

Stichwörter der Autor:innen

contrastive learningdownstream analysisspatial clusteringspatial transcriptomicstrajectoryvariational graph auto-encoder

Autor:innen und Institutionen

  • School of Automation Science and Engineering, Xi'an Jiaotong University, No. 28 Xianning West Road, Beilin District, Xi'an, Shannxi, 710049, China.
  • Research Institute of Xi'an Jiaotong University, Zhejiang, No. 27 Ningtai Road, Xiaoshan District, Hangzhou, Zhejiang, 311200, China.
  • Sichuan Digital Economy Industry Development Research Institute, No. 999 Xihua Avenue, Jinniu District, Chengdu, Sichuan, 610036, China.
  • College of Information Engineering, Northwest A&F University, No. 22, Xinong Road, Yangling District, Xianyang, Shaanxi, 712100, China.
  • Department of Chromosome Biology, Max Planck Institute for Plant Breeding Research, Carl-von-Linné-Weg 10, Cologne North Rhine Westphalia, Cologne, 50829, Germany.

ORCID-Profile: Laiyi Fu, Danyang Wu, Hequan Sun

Förderung

  • National Natural Science Foundation of China
  • Natural Science Basic Research Plan in Shaanxi Province of China
  • China Postdoctoral Science Foundation
  • Postdoctoral Research Project in Shaanxi Province
  • Zhejiang Provincial Natural Science Foundation of China
  • Fundamental Research Funds for the Central Universities
  • Sichuan Science and Technology Program

Zitieren

APA

Fu, L., Cui, W., Chen, Y., Wu, D., & Sun, H. (2026). Biased multi-view contrastive learning with attentive masking for spatial transcriptomic analysis. Briefings in Bioinformatics, 27(3). https://doi.org/10.1093/bib/bbag215

Vancouver

Fu L, Cui W, Chen Y, Wu D, Sun H. Biased multi-view contrastive learning with attentive masking for spatial transcriptomic analysis. Briefings in Bioinformatics. 2026;27(3). doi:10.1093/bib/bbag215

BibTeX
@article{Fu2026,
  author = {Fu, Laiyi and Cui, Wenkai and Chen, Yifan and Wu, Danyang and Sun, Hequan},
  title = {Biased multi-view contrastive learning with attentive masking for spatial transcriptomic analysis},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  volume = {27},
  number = {3},
  doi = {10.1093/bib/bbag215}
}
Kostenlos starten →