Paper Review

The field of spatial transcriptomics has made significant strides with the advent of high-resolution technologies, enabling the generation of large-scale multi-sample datasets. However, most existing analytical frameworks are primarily designed for single-sample analysis and lack the flexibility, scalability, and advanced capabilities needed for comprehensive multi-sample studies.

Recently, A paper published in Nature communications, titled "Stereopy: modeling comparative and spatiotemporal cellular heterogeneity via multi-sample spatial transcriptomics" addresses this gap, introducing a novel framework tailored to handle the complexities of multi-sample spatial transcriptomics. Stereopy features a scalable architecture with a universal data container (MsData) for multimodal data storage, a scope controller (MSS) for subset analysis, and an integrative transformer for seamless data processing. This design extends the AnnData format, ensuring compatibility and scalability for multi-sample datasets.

The three advanced algorithms in Stereopy have enabled significant advancements in comparative analysis, temporal analysis, and 3D spatial multi-sample omics analysis. The Multi-Sample Cell Community Detection (CCD) Algorithm enables the identification of common or specific cell communities across case-control samples, revealing global and local diversities in cell types and their co-occurrence. The Spatially Resolved Temporal Gene Pattern Inference (TGPI) Algorithm integrates spatial and temporal data to detect spatiotemporal gene patterns, providing insights into molecular dynamics during development. Furthermore, the NicheReg3D Tool reconstructs 3D cell niches and infers cell-gene interaction networks, thereby linking intercellular communications with intracellular regulatory mechanisms within spatial constraints.

In this study, Stereo-seq's high-resolution data were of importance in Stereopy's advanced analyses. For example, in the temporal analysis of mouse embryos, Stereo-seq enabled the detection of spatiotemporal gene patterns, while in 3D studies, its detailed spatial information facilitated the identification of cell-niche interactions and comprehensive ligand-receptor pairs in the embryonic heart.

Find out more about the research: https://www.nature.com/articles/s41467-025-58079-9#citeas

About STOmics' Stereo-seq:
STOmics offers the most advanced spatiotemporal multi-omics technology, enabling unbiased discovery to answer biological questions in scientific research and clinical applications. Currently, we offer spatial transcriptomics solutions, including Stereo-seq v1.3 for fresh frozen samples, Stereo-seq OMNI for FFPE samples, Stereo-seq Large Chip Designs (LCD) for centimeter-level fresh frozen samples (now up to 2cm x 3cm), and a spatial multi-omics solution - Stereo-CITE for high-plex spatial proteo-transcriptome co-detection. https://en.stomics.tech/