Abstract 2442: STHD: Probabilistic cell typing of single spots in whole transcriptome spatial data with high definition

Abstract Understanding of the cellular heterogeneity of tumor has been essential for uncovering cancer progression mechanisms and developing effective cancer immunotherapies. Emerging spatial transcriptomics (ST) technologies that enable single- and sub-cellular resolution profiling of gene expression across the whole transcriptome have recently become compatible with cancer samples. However, the transition to high-definition ST significantly increased data sparsity and dimensionality, posing significant computational challenges. For example, identifying cell types, deciphering neighborhood structure, and detecting differential expression are all crucial steps to interpret cancer ST samples. Here we present STHD, a novel machine learning method for probabilistic cell typing of single spots in whole-transcriptome, high-resolution ST data. Unlike the current binning-aggregation-deconvolution strategy, STHD directly models gene expression at single-spot level to infer cell type identities without cell segmentation or spot aggregation. STHD addresses sparsity by modeling count statistics, incorporating neighbor similarities, and leveraging reference single-cell RNA-seq data. We show in VisiumHD data that STHD accurately predicts cell type identities at single-spot level, which achieves precise segmentation of both global tissue architecture and local multicellular neighborhoods. The high-resolution labels facilitate various downstream analyses, including cell type-stratified bin aggregation, spatial compositional comparisons, and cell type-specific differential expression analyses. We applied STHD to analyze the newest colon cancer VisiumHD data, which identified patterns like spatial regions of cancer cell-specific stemness, spatially heterogenous cytokine communication activities, confined hypoxia macrophages, and distinct T cell-myeloid cell interactions and cytotoxicity activities located in different immune hubs. Moreover, STHD labels further reveal frontlines of inter-cell type interactions at immune hubs. STHD is scalable and generalizable across diverse samples, tissues, diseases, and cancer types, facilitating genome-wide analyses in various spatial organization contexts. Overall, STHD maps cell type of cancer ST samples with ultra-resolution and facilitates discovery of novel spatial biomarkers and molecular mechanism of cancer progression and treatment responses. The STHD software is publicly available at https://github.com/yi-zhang-compbio-lab/STHD. Citation Format: Chuhanwen Sun, Yi Zhang. STHD: Probabilistic cell typing of single spots in whole transcriptome spatial data with high definition [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2442.