Inferring single-cell resolution spatial gene expression via fusing spot-based spatial transcriptomics, location and histology using GCN

Abstract Spatial transcriptomics technology allows for the detection of cellular transcriptome information while preserving the spatial location of cells. This capability enables researchers to better understand the cellular heterogeneity, spatial organization and functional interactions in complex biological systems. However, current technological methods are limited by low resolution, which reduces the accuracy of gene expression levels. Here, we propose scstGCN, a multimodal information fusion method based on Vision Transformer (ViT) and Graph Convolutional Network (GCN) that integrates histological images, spot-based spatial transcriptomics data and spatial location information to infer super-resolution gene expression profiles at single-cell level. We evaluated the accuracy of the super-resolution gene expression profiles generated on diverse tissue ST datasets with disease and healthy by scstGCN along with their performance in identifying spatial patterns, conducting functional enrichment analysis, and tissue annotation. The results show that scstGCN can predict super-resolution gene expression accurately, aid researchers in discovering biologically meaningful differentially expressed genes and pathways. Additionally, scstGCN can segment and annotate tissues at a finer granularity, with results demonstrating strong consistency with coarse manual annotations. Key Points scstGCN combines multi-modal information including histology image, spot-based spatial transcriptomics (ST) data, and physical spatial location through deep learning methods to achieve single-cell resolution of spot-based ST data without requiring single-cell references. scstGCN employs GCN to capture complex relationships between neighboring cells, facilitating the integration of multimodal feature information based on single-cell level, and then accurately infers single-cell resolution spatial gene expression. scstGCN can infer single-cell resolution gene expression across the entire tissue region. Through transfer learning, gene expression in three-dimensional tissues can be characterized efficiently. Furthermore, it demonstrates outstanding performance in spatial patterns enhancement, functional enrichment analysis, and annotate tissues at the high-resolution.