Deciphering cell–cell communication in spatial transcriptomics through single-cell-informed graph convolutional networks

Pengfei Gao

doi:10.1016/j.patcog.2026.115096

Deciphering cell–cell communication in spatial transcriptomics through single-cell-informed graph convolutional networks

2026-10-05

0

PRE

AI

DOI：10.1016/j.patcog.2026.115096

Original

Original request for help

Save

Abstract

En 中文

• We propose SMGN, a unified GCN framework for spatial cell–cell communication inference. • We introduce a pseudo-spot strategy to address intra-spot heterogeneity in spatial transcriptomics. • We design a Gaussian spatial tensor to model ligand diffusion and communication range. • We develop a communication-aware gene expression prediction module for validation.

Keywords:

Cell-cell communication

Cell-type deconvolution

Graph neural network

Spatial transcriptomics

Single-cell RNA sequencing

Deciphering cell–cell communication in spatial transcriptomics through single-cell-informed graph convolutional networks

Abstract

Journal

Organization

Cited Papers