Abstract

Imaging based spatial transcriptomics (iST), such as MERFISH, CosMx SMI, and Xenium, quantify gene expression level across cells in space, but more importantly, they directly reveal the subcellular distribution of RNA transcripts at the single-molecule resolution. The subcellular localization of RNA molecules plays a crucial role in the compartmentalization-dependent regulation of genes within individual cells. Understanding the intracellular spatial distribution of RNA for a particular cell type thus not only improves the characterization of cell identity but also is of paramount importance in elucidating unique subcellular regulatory mechanisms specific to the cell type. However, current cell type annotation approaches of iST primarily utilize gene expression information while neglecting the spatial distribution of RNAs within cells. In this work, we introduce a semi-supervised graph contrastive learning method called Focus, the first method, to the best of our knowledge, that explicitly models RNAs subcellular distribution and community to improve cell type annotation. Focus first constructs gene neighborhood networks based on the subcellular colocalization relationship of RNA transcripts. Next, the subcellular graph of each cell can be augmented by adding important edges and nodes or removing trivial edges and nodes. Focus then aims to maximize the similarity between positive pairs from two augmented views of the same cell and minimize the similarity between negative pairs from different cells within a common batch. Guided by a limited amount of labeled data, Focus is capable of assigning cell type identities for the entire datasets at high accuracy. Extensive experiments demonstrate the effectiveness of Focus compared to existing state-of-the-art approaches across a range of spatial transcriptomics platforms and biological systems. Furthermore, Focus enjoys the advantages of revealing intricate cell type-specific subcellular spatial gene patterns and providing interpretable subcellular gene analysis, such as defining the gene importance score. Importantly, with the importance score, Focus identifies genes harboring strong relevance to cell type-specific pathways, indicating its potential in uncovering novel regulatory programs across numerous biological systems. Focus is freely accessible at https://github.com/OmicsML/focus.