EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MAGNET: Multi-view graph autoencoder with cell-gene attention for cell interaction network reconstruction from spatial transcriptomics

Chengxu Han iD, Zhangdi Song, Zimu Xu, Jiaxing Chen iD

PLOS Computational Biology · 2025

DOI: 10.1371/journal.pcbi.1013810 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Computational Biology, 2025. Verfasst von Chengxu Han und 3 weiteren Autor:innen. Beteiligte Einrichtung: Guangdong Provincial/Zhuhai Key Laboratory IRADS, Beijing Normal University-Hong Kong Baptist University United International College, Zhuhai, China. Hauptthemen laut MEDLINE: Cell Communication, Gene Regulatory Networks, Transcriptome, Gene Expression Profiling. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, BNBU Research Grant, BNBU Start-up Fund.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurately inferring cell-cell interactions from spatial transcriptomics data remains challenging due to tissue complexity and spatial heterogeneity. Recent deep learning models have started to combine different types of cellular information, such as spatial proximity, gene expression similarity, ligand–receptor signaling, and in some cases, gene regulatory networks. However, they often treat a cell’s external interactions and internal gene regulation separately, merging them at the final step using simple concatenation or addition. This limits the model’s ability to capture how cell–cell communication and internal molecular states are connected. In this paper, we present MAGNET (Multi-view Graph Autoencoder with Cell-Gene Attention Network), a framework that reconstructs cell–cell interaction networks by building multiple biological graphs and developing a Cell-Gene attention module to link a cell’s environment with its internal gene activity in a unified representation. On benchmark datasets (seqFISH, MERFISH, STARMAP), MAGNET demonstrates superior performance in reconstructing cell-cell interaction networks, achieving an Average Precision(AP) of 0.901 on the seqFISH dataset and outperforming TENET by 0.185. The Cell-Gene Attention module is critical to MAGNET’s performance, as its removal alone caused the AP on the seqFISH dataset to drop from 0.901 to 0.521. Applied to a breast cancer dataset, MAGNET found functional heterogeneity among cancer cells, distinguishing clusters with molecular signatures for either immune evasion or autonomous tumor growth.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Cell CommunicationGene Regulatory NetworksTranscriptomeGene Expression ProfilingHumansComputational BiologyBreast NeoplasmsAlgorithmsFemaleDeep LearningAutoencoder

Autor:innen und Institutionen

  • Guangdong Provincial/Zhuhai Key Laboratory IRADS, Beijing Normal University-Hong Kong Baptist University United International College, Zhuhai, China.
  • Department of Computer Science, Hong Kong Baptist University, Hong Kong SAR, China.
  • Institute of Computer Science, University of Bern, Bern, Switzerland.
  • Key Laboratory for Artificial Intelligence and Multi-Model Data Processing, Department of Education of Guangdong Province, Zhuhai, China.

ORCID-Profile: Chengxu Han, Jiaxing Chen

Förderung

  • National Natural Science Foundation of China
  • BNBU Research Grant
  • BNBU Start-up Fund

Zitieren

APA

Han, C., Song, Z., Xu, Z., & Chen, J. (2025). MAGNET: Multi-view graph autoencoder with cell-gene attention for cell interaction network reconstruction from spatial transcriptomics. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1013810

Vancouver

Han C, Song Z, Xu Z, Chen J. MAGNET: Multi-view graph autoencoder with cell-gene attention for cell interaction network reconstruction from spatial transcriptomics. PLOS Computational Biology. 2025. doi:10.1371/journal.pcbi.1013810

BibTeX
@article{Han2025,
  author = {Han, Chengxu and Song, Zhangdi and Xu, Zimu and Chen, Jiaxing},
  title = {MAGNET: Multi-view graph autoencoder with cell-gene attention for cell interaction network reconstruction from spatial transcriptomics},
  journal = {PLOS Computational Biology},
  year = {2025},
  doi = {10.1371/journal.pcbi.1013810}
}
Kostenlos starten →