Choosing the Right Spatial Transcriptomics Platform for Your Research

Understanding where genes are expressed within tissues can be just as important as knowing which genes are expressed. Tissue development, immune responses, and disease progression all depend on the spatial organization of cells. Traditional single-cell RNA sequencing captures gene expression but requires tissue dissociation, losing that spatial context.

Spatial transcriptomics addresses this limitation by measuring gene expression directly within intact tissue, preserving spatial context and tissue architecture ¹. This lets researchers study cellular neighborhoods, tissue organization, and cell-cell interactions that drive physiology and disease.

Current spatial transcriptomics platforms differ in spatial resolution, transcriptome coverage, throughput, and analytical requirements. Rather than asking ” Which platform is the best?”, a better question is:

Which platform best answers my biological question?

This article compares the major approaches and the key considerations for choosing between them.

Why Choosing the Right Platform Matters

Platform choice shapes every stage of a study, from experimental design to computational analysis and interpretation.

Before choosing a technology, consider:

• Are you exploring tissue architecture across an entire tissue section?
• Do you need to identify rare cell populations at single-cell resolution?
• Are you discovering new biomarkers or validating a predefined gene panel?
• Will you analyze fresh frozen tissue, FFPE clinical samples, or both?
• How large is the tissue section, and how many samples will you process?

Weighing these trade-offs upfront saves time and resources, and ensures the data you generate actually answers your question.

Two Major Approaches to Spatial Transcriptomics

Although many commercial platforms are available, most spatial transcriptomics technologies fall into two broad categories: sequencing-based and imaging-based approaches ². While both approaches measure spatial gene expression, they differ fundamentally in how transcripts are captured: sequencing-based methods generate sequencing libraries from spatially indexed RNA, whereas imaging-based methods directly visualize RNA molecules within intact tissue.

1. Sequencing-Based Spatial Transcriptomics

Sequencing-based approaches capture RNA—or probe-derived molecules in FFPE workflows—from spatially indexed locations, then convert them into sequencing libraries that preserve spatial coordinates for reconstructing gene expression across the tissue.

Strengths

• Whole-transcriptome profiling
• Large tissue coverage
• High throughput for discovery studies
• Well suited for tissue architecture and biomarker discovery

Considerations

• Spatial resolution varies between platforms
• Individual cells may occupy multiple capture spots, requiring computational methods for cell-type deconvolution or integration with single-cell reference datasets
• Performance and sample compatibility differ between fresh frozen and FFPE tissues depending on the platform

2. Imaging-Based Spatial Transcriptomics

Imaging-based technologies detect RNA directly within intact tissue using highly multiplexed in situ hybridization and fluorescence imaging, localizing transcripts to segmented cells at single-cell or subcellular resolution ³.

Strengths

• Single-cell or subcellular spatial resolution
• Excellent for studying cellular interactions and tissue microenvironments
• Direct visualization of transcript localization
• Strong performance for analyzing predefined biomarkers

Considerations

• Most platforms still rely on predefined gene panels (size varies significantly), though whole-transcriptome imaging is emerging—e.g., CosMx WTx or Atera (~19,000 genes)
• Accurate cell segmentation is critical for downstream analyses and biological interpretation
• Imaging experiments typically generate large datasets that require substantial computational resources

Comparing Key Features

Each technology trades off spatial resolution, transcriptome coverage, throughput, sample compatibility, and computational complexity.

Feature	Sequencing-Based	Imaging-Based
Representative platforms	Visium, Visium HD, GeoMx*	Xenium, CosMx, MERSCOPE, Atera
Spatial resolution	Spot to near single-cell (platform dependent)	Single-cell to subcellular
Gene coverage	Whole transcriptome	Mostly targeted panels (varies by platform); a few (e.g., CosMx WTx, Atera) now offer whole-transcriptome coverage
Tissue coverage	Large tissue regions	Typically smaller imaging areas
Throughput	High	Moderate
Cell segmentation	Optional	Essential
FFPE compatibility	Platform dependent	Commonly supported by several platforms
Typical output	Gene expression matrix with spatial coordinates	Transcript coordinates, cell segmentation, morphology images
Computational complexity	Moderate	Higher due to image processing and segmentation
Best suited for	Tissue architecture, biomarker discovery, large-scale profiling	Cell-cell interactions, spatial organization, high-resolution cellular analysis

*GeoMx performs region-of-interest–based spatial profiling rather than whole-slide spatial transcriptomics and is often considered a complementary spatial profiling technology. Atera is a newly launched (2026) imaging-based platform ⁴ from 10x Genomics offering whole-transcriptome coverage (~19,000 genes) with single-cell resolution at higher throughput than earlier imaging platforms.

Sequencing-based platforms prioritize breadth: whole-transcriptome coverage across large tissue areas at high throughput, but at spot-to-near-single-cell resolution, making them best for tissue architecture mapping and large-scale biomarker discovery. 

On the other hand, imaging-based platforms prioritize precision: single-cell to subcellular resolution with essential cell segmentation, typically over smaller tissue areas and targeted gene panels (though CosMx WTx and Atera now offer whole-transcriptome options), making them best for studying cell-cell interactions and fine-grained spatial organization. 

Figure 1. Comparison of sequencing-based and imaging-based spatial transcriptomics technologies. Source: https://geneviatechnologies.com/bioinformatics-analyses/spatial-transcriptomic-data-analysis/

Figure 2. Example of sequencing-based (Visium HD) versus imaging-based (Xenium) spatial transcriptomics data on the same tissue type, illustrating the whole-transcriptome/spot-based versus targeted/single-cell tradeoff described above. Image courtesy of 10x Genomics.

Choosing a Platform Based on Your Biological Question

Start with the biological question you want to answer, rather than technical specifications alone.

If your goal is to discover novel biomarkers or profiling transcriptional programs tissue-wide, whole-transcriptome spatial profiling is generally preferred because it avoids predefined targets. This can now be achieved using both sequencing-based platforms (e.g., Visium HD, Stereo-seq) and emerging imaging-based platforms (e.g., CosMx WTx and Atera), each with different trade-offs in throughput, field of view, and resolution.

If you want to investigate cell-cell interactions, cellular neighborhoods, or rare cell populations, imaging-based technologies offer the spatial precision to resolve individual cells and their microenvironments.

Figure 3. Decision tree for selecting a spatial transcriptomics technology based on the research objective. Protein-based spatial profiling platforms (e.g., CODEX and PhenoCycler) are included as complementary technologies for protein biomarker validation where appropriate.

Beyond the Experiment: Computational Considerations

Choosing a platform also determines the downstream computational workflow.

Typical steps include image registration, quality control, batch correction, normalization, cell type annotation, differential expression, spatial domain identification, neighborhood analysis, and cell-cell communication analysis. Imaging data additionally require cell segmentation and transcript assignment; sequencing data often benefit from integration with single-cell RNA-seq references.

Planning for these requirements early helps you pick both the right technology and the right analysis tools.

What Data Will You Receive?

Output formats vary, but most spatial transcriptomics datasets include:

• Tissue morphology images (H&E, immunofluorescence, or brightfield)
• Gene expression matrices
• Spatial coordinates
• Cell segmentation masks (primarily for imaging-based platforms)
• Metadata for image alignment and spatial scaling

Together, these data support the downstream analyses described above, plus tasks like trajectory inference.

Explore Public Spatial Transcriptomics Datasets Before Designing Your Study

One of the best ways to learn a platform’s strengths and limitations is to explore public datasets before running your own experiment.

BioTuring’s SpatialX provides access to public datasets across both sequencing- and imaging-based platforms. Exploring them within one consistent workflow lets you compare platform characteristics and downstream analyses before committing to a technology.

Whether you’re examining tissue architecture with Visium, single-cell interactions with Xenium, or comparing across technologies, exploring public data first can save you from costly missteps.

Figure 4: Selecting available demo datasets in SpatialX (upper) or querying BioTuring’s collection of published spatial transcriptomics studies (below) through Talk2Data.

Not sure which platform fits your study? Request a demo and explore public spatial datasets with our team. 

Request a Demo → 

Conclusion

Spatial transcriptomics technologies are complementary, not competing. Sequencing-based approaches excel at large-scale, genome-wide profiling; imaging-based methods offer the precision to resolve cellular interactions at single-cell resolution.

The right question isn’t which platform is most advanced, but which best answers your biological question—weighing resolution, transcriptome coverage, sample compatibility, tissue size, throughput, and analytical needs.

Matching technology to objective—and exploring public datasets first—helps you design experiments that maximize both insight and efficiency.

References

1. Ståhl, P. L. et al. Visualization and analysis of gene expression in tissue sections by spatial transcriptomics. Science (1979). 353, 78–82 (2016).

2. Wang, Y. et al. Spatial transcriptomics: Technologies, applications and experimental considerations. Genomics 115, 110671 (2023).

3. Chen, K. H., Boettiger, A. N., Moffitt, J. R., Wang, S. & Zhuang, X. Spatially resolved, highly multiplexed RNA profiling in single cells. Science (1979). 348, (2015).

4. Janesick, A., Toh, M., Mohabbat, S. & Kravitz, S. Abstract 6216: Novel whole transcriptome spatial transcriptomics technology reveals CAF/TAM-mediated basement membrane remodeling at the invasive front. Cancer Res. 86, 6216–6216 (2026).