Abstract:The spatial arrangement of cells is vital in developmental processes and organogenesis in multicellular life forms. Deep learning models trained with spatial omics data uncover complex patterns and relationships among cells, genes, and proteins in a high-dimensional space, providing new insights into biological processes and diseases. State-of-the-art in silico spatial multi-cell gene expression methods using histological images of tissue stained with hematoxylin and eosin (H&E) to characterize cellular heterogeneity. These computational techniques offer the advantage of analyzing vast amounts of spatial data in a scalable and automated manner, thereby accelerating scientific discovery and enabling more precise medical diagnostics and treatments. In this work, we developed a vision transformer (ViT) framework to map histological signatures to spatial single-cell transcriptomic signatures, named SPiRiT (Spatial Omics Prediction and Reproducibility integrated Transformer). Our framework was enhanced by integrating cross validation with model interpretation during hyper-parameter tuning. SPiRiT predicts single-cell spatial gene expression using the matched histopathological image tiles of human breast cancer and whole mouse pup, evaluated by Xenium (10x Genomics) datasets. Furthermore, ViT model interpretation reveals the high-resolution, high attention area (HAR) that the ViT model uses to predict the gene expression, including marker genes for invasive cancer cells (FASN), stromal cells (POSTN), and lymphocytes (IL7R). In an apple-to-apple comparison with the ST-Net Convolutional Neural Network algorithm, SPiRiT improved predictive accuracy by 40% using human breast cancer Visium (10x Genomics) dataset. Cancer biomarker gene prediction and expression level are highly consistent with the tumor region annotation. In summary, our work highlights the feasibility to infer spatial single-cell gene expression using tissue morphology in multiple-species, i.e., human and mouse, and multi-organs, i.e., mouse whole body morphology. Importantly, incorporating model interpretation and vision transformer is expected to serve as a general-purpose framework for spatial transcriptomics.

A framework for gene representation on spatial transcriptomics

stFormer: a foundation model for spatial transcriptomics

Spanve: an Statistical Method to Detect Clustering-friendly Spatially Variable Genes in Large-scale Spatial Transcriptomics Data

HyperGCN: an effective deep representation learning framework for the integrative analysis of spatial transcriptomics data

A contrastive learning approach to integrate spatial transcriptomics and histological images

Spatial transcriptomics: Technologies, applications and experimental considerations

Analysis and visualization of spatial transcriptomic data

A Unified Probabilistic Framework for Modeling and Inferring Spatial Transcriptomic Data

Imputing spatial transcriptomics through gene network constructed from protein language model

Gene representation bias in spatial transcriptomics

Inferring single-cell spatial gene expression with tissue morphology via explainable deep learning

stEnTrans: Transformer-based deep learning for spatial transcriptomics enhancement

Cross-expression analysis reveals patterns of coordinated gene expression in spatial transcriptomics

A multi-modality and multi-granularity collaborative learning framework for identifying spatial domains and spatially variable genes

SpatialMap: Spatial Mapping of Unmeasured Gene Expression Profiles in Spatial Transcriptomic Data Using Generalized Linear Spatial Models

A Guidebook of Spatial Transcriptomic Technologies, Data Resources and Analysis Approaches

Integrating spatial and single-cell transcriptomics data using deep generative models with SpatialScope

Spatially aware self-representation learning for tissue structure characterization and spatial functional genes identification

Advances in spatial transcriptomic data analysis

Embryo-scale, single-cell spatial transcriptomics