Deciphering spatial domains from spatial multi-omics with SpatialGlue

Abstract

Integration of multiple data modalities in a spatially informed manner remains an unmet need for exploiting spatial multi-omics data. Here, we introduce SpatialGlue, a novel graph neural network with dual-attention mechanism, to decipher spatial domains by capturing the significance of each modality and neighbor graph in cross-omics and intra-omics integration. We demonstrate that SpatialGlue can accurately aggregate cell types into spatial domains at a higher resolution across different tissue types and technology platforms, as well as gain biological insights into cross-modality spatial correlations.

Spatially Resolved Gene Expression is Not Necessary for Identifying Spatial Domains

“…For instance, in addition to the two primary modalities of SRT data (i.e., gene expression and spatial coordinates) as commonly done by most methods [9][10][11][12] , new methods integrate supplementary images (e.g., immunohistochemistry images 13 , H&E images 14,15 , Chromatin images 16 , etc. ), human annotations 17 , paired scRNA-seq data 18 or other-spatial-omics data 19 . In this context, we explore in an opposite direction, investigating whether it is possible to accurately identify spatial domains with less information.…”