EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

Multi-Scale Graph Attention for Cell-Cell Interaction Inference in Spatial Transcriptomics

Ángela González, Alejandro Iglesias

Journal of geospace science · 2026-08-01 · 6 (2) · 72-82

DOI: 10.71465/jgs68 ↗

At a glance

Journal article in Journal of geospace science, published online on 2026-08-01. Written by Ángela González and 1 co-authors. The full text is freely available (open access, CC BY 4.0).

AbstractBibliografiske oplysningerCiteRelated studies

Abstract (original language)

The rapid advancement of spatial transcriptomics has provided unprecedented opportunities to investigate the spatial organization and communication of cells within complex tissues. While traditional single-cell RNA sequencing reveals transcriptomic heterogeneity, it fundamentally lacks the spatial context necessary to understand how cells physically and biochemically interact in their native microenvironments. In response to this limitation, computational methods utilizing graph neural networks have emerged as powerful tools for deciphering cell-cell interactions from spatial transcriptomics data. This paper presents a comprehensive theoretical framework and computational methodology based on multi-scale graph attention networks for inferring cell-cell interactions. By representing tissue structures as spatial graphs where nodes correspond to cells or capture spots and edges denote spatial proximity, the proposed framework captures both local microenvironmental contexts and broader structural patterns. The multi-scale approach leverages varying receptive fields to aggregate features from neighboring cells at different distances, enabling the identification of both short-range juxtacrine signaling and long-range paracrine communication. Furthermore, the integration of attention mechanisms allows the model to dynamically weight the importance of neighboring nodes based on their transcriptomic profiles, mitigating the noise inherent in spatial data. Through extensive theoretical analysis and methodological description, this study demonstrates how multi-scale graph attention can significantly improve the accuracy and robustness of interaction inference, ultimately providing deeper biological insights into tissue development, immune responses, and disease pathogenesis.

Cite

APA

González, Á., & Iglesias, A. (2026). Multi-Scale Graph Attention for Cell-Cell Interaction Inference in Spatial Transcriptomics. Journal of geospace science, 6(2), 72-82. https://doi.org/10.71465/jgs68

Vancouver

González Á, Iglesias A. Multi-Scale Graph Attention for Cell-Cell Interaction Inference in Spatial Transcriptomics. Journal of geospace science. 2026;6(2):72-82. doi:10.71465/jgs68

BibTeX
@article{Gonzlez2026,
  author = {González, Ángela and Iglesias, Alejandro},
  title = {Multi-Scale Graph Attention for Cell-Cell Interaction Inference in Spatial Transcriptomics},
  journal = {Journal of geospace science},
  year = {2026},
  volume = {6},
  number = {2},
  pages = {72--82},
  doi = {10.71465/jgs68}
}
Start for free →