What Is Spatial Transcriptomics and When Do You Need It?

• Spatial resolution matters most when cell-cell proximity, tissue architecture, or microenvironment context drives the biology, from tumor margins to immune niches.
• Recent peer-reviewed studies show spatial context revealing cell states and interactions that single-cell RNA sequencing (scRNA-seq) alone missed.
• iSCanGuide™ is the only guided, GPU-accelerated platform that unifies single-cell and spatial transcriptomics with direct pathway-level analysis, making spatial biology accessible without a dedicated computational spatial biology team.

Spatial transcriptomics refers to a set of technologies that measure gene expression while preserving the physical location of that expression within intact tissue. Rather than a dissociated cell suspension, the input is a tissue section. The output of these experiments links gene expression data back to defined coordinates on that section.

Two broad technical approaches dominate the field. Imaging-based methods use targeted in situ hybridization or sequencing to directly visualize transcripts within cells. Sequencing-based methods capture RNA from spatially barcoded spots or beads arrayed across a tissue section, then computationally map the resulting expression profiles back to their tissue location (Table 1). How spatial transcriptomics works in practice depends on which approach a given platform uses, but the shared principle is the same: gene expression data that comes with a map attached.

Table 1. Spatial transcriptomics: Two technical approaches.

The field has moved quickly. A widely cited historical review traces spatial transcriptomics methods back to low-throughput in situ approaches from the late 1980s, through the technological expansion that produced today’s high-throughput platforms.¹ Many current single-cell spatial transcriptomics platforms now achieve single-cell, not just spot-level spatial resolution, closing much of the gap between “spatial” and “single-cell” as separate categories.

A UMAP plot, the standard way to visualize single-cell clusters, tells you which cells are transcriptionally similar to each other. It does not tell you whether those cells were ever near each other in the tissue, however, that distinction turns out to matter a great deal in practice.

In oncology, spatial studies of non-small cell lung cancer have mapped specific fibroblast and macrophage populations into defined tissue neighborhoods relative to tumor cells, revealing organizational patterns in the tumor microenvironment that dissociated scRNA-seq data cannot capture on its own.² Related work in breast cancer has used spatial mapping to characterize how immunosuppressive fibroblast states are distributed across tissue and how their position relates to immune exclusion from the tumor.³ In both cases, the cell types were identifiable from expression alone, but the biological story, and its therapeutic implications, depended on where those cells were located in the tissue.

The same principle holds beyond cancer research. A 2024 review in the Journal of Allergy and Clinical Immunology describes how spatial transcriptomics now maps fine-grained immune cell states to precise tissue locations along with their surrounding cellular context, allowing researchers to study immune niches — functionally organized clusters of interacting immune and stromal cells — rather than treating immune cell counts in isolation.⁴ A more recent review extends this further, describing how spatial approaches are being used to characterize B and T cell clonal niches directly within tissue architecture.⁵

Whether the question is how tumor and immune cells organize at an invasive margin, or how immune niches are structured in healthy versus diseased tissue, the answer depends on spatial position — something a UMAP alone cannot supply.

Spatial analysis has historically been harder for labs to adopt than single-cell analysis alone, largely because the computational tooling has been fragmented, compute-intensive, and dependent on specialized expertise.

That narrative is now changing. Integrated analytical workflows now make it possible to connect single-cell and spatial analysis in one environment instead of stitching together separate point solutions for clustering, spatial domain detection, and downstream interpretation. iSCanGuide™ is the platform setting this standard: it’s the only guided, GPU-accelerated environment that unifies single-cell and spatial transcriptomics with direct integration into validated mechanistic pathway analysis. Spatial context is preserved through every downstream step, including cell-cell communication analysis and pathway-level interpretation via iPathwayGuide™’s validated pathway analysis engine, built on nearly 20 years of peer-reviewed methodology.

Figure 1. Converging lines of evidence for epithelial identity in a Visium HD colorectal cancer section, analyzed end-to-end in iSCanGuide™. The same Visium HD dataset (sample P2-CRC) is shown as a UMAP embedding (left column) and in tissue space over the H&E image (right column). (A, B) Expression of the canonical epithelial marker EPCAM, plotted on the embedding and in situ; high-EPCAM bins trace the glandular and tumor epithelium. (C, D) Unsupervised clustering (Clusters) resolves 16 transcriptional clusters, several of which occupy the EPCAM-high territory in both views. (E, F) Reference-based annotation with SingleR (BlueprintENCODE reference) labels those same regions as epithelial cells (red); all other predicted cell types are shown in grey. Clusters 1, 2, 3, 5, 9, 10 and 11 map to the epithelial compartment. The agreement across marker expression, unsupervised structure, and reference-based labels, visible simultaneously in embedding and tissue space, is what lets iSCanGuide™ assign cell identity with confidence rather than from a single line of evidence. Scale bars, 2 mm. (Adapted from 10X., 2024)⁶

Spatial transcriptomics is moving from a specialized add-on to a standard layer of transcriptomic analysis, particularly in oncology, immunology, and developmental biology. As it becomes more accessible, the questions it can answer are expanding too: how cell-cell communication shifts across tissue regions, how therapeutic response correlates with spatial organization, how disease progression reorganizes tissue architecture over time. To delve further into why spatial context changes biological conclusions, and what to look for when deciding whether a project needs spatial analysis, visit our spatial learning center or reach out to our team to discuss solutions that can help you achieve your research goals.

Unlike standard scRNA-seq, spatial transcriptomics preserves the link between transcription and location within a cell and intact tissue. Studies focused on the tissue microenvironment and architecture, cell type identification, cell-cell communication, biomarker identification, and therapeutic response all benefit from combined information on cell location and transcriptional activity. These data can help resolve mechanisms that drive processes such as disease progression or immune response.

iSCanGuide™ offers deep biological interpretation of spatial transcriptomics data that includes:

• Spatial domain detection
• Tissue architecture mapping
• Spatially variable genes
• Cell localization
• trajectory inference
• Cell-type identification
• Rapid real-time manipulation of millions of cell datasets

Yes. Advaita Bio’s iSCanGuide™ platform for single-cell and spatial analysis integrates seamlessly with iPathwayGuide™, our analytical platform built for pathway-level biology.

Reviewed by Beverly V. Robinson, PhD (Genetics & Molecular Biology, Emory University), Bioinformatics Scientist at Advaita