Spatial Transcriptomics Technologies

  • Spatial transcriptomics technologies are experimental and computational methods used to measure gene expression while preserving information about the physical location of RNA molecules within tissues. Unlike conventional bulk RNA sequencing, which measures gene expression across a mixture of cells, and conventional single-cell RNA sequencing (scRNA-seq), which generally requires tissue dissociation, spatial transcriptomics maintains the relationship between molecular activity and tissue architecture. These technologies enable researchers to investigate which genes are expressed, where their expression occurs, which cell populations occupy particular tissue regions, and how cells are organized within their biological environments. Spatial transcriptomics has therefore become an important area of bioinformatics, molecular biology, developmental biology, cancer research, and precision medicine.
  • The fundamental principle of spatial transcriptomics is to associate RNA molecules with identifiable positions in a tissue section. This is achieved through spatial barcoding, molecular capture, fluorescence-based imaging, in situ hybridization, or sequencing directly within tissue. Depending on the technology, measurements may represent groups of cells, individual cells, or subcellular locations. Some platforms measure thousands of genes across relatively large tissue areas, whereas others detect selected RNA targets at high spatial resolution. The choice of technology depends on the research objective, tissue type, required spatial resolution, transcriptome coverage, sensitivity, throughput, and available resources.
  • Spatial transcriptomics technologies can broadly be divided into two major categories: sequencing-based methods and imaging-based methods. Sequencing-based technologies generally capture RNA molecules using spatially indexed probes and identify the captured transcripts through next-generation sequencing (NGS). Imaging-based technologies detect specific RNA molecules directly within tissue sections using fluorescent probes and repeated imaging cycles or other molecular detection systems. A third category includes in situ sequencing approaches, which identify RNA sequences directly within their original tissue environment. These categories overlap in some technologies, but they provide a useful framework for understanding the main experimental principles.
  • Sequencing-based spatial transcriptomics uses spatially barcoded capture probes to retain information about the location of RNA molecules. A tissue section is placed on a slide containing capture regions with known spatial identifiers. RNA molecules released or made accessible from the tissue bind to these probes, and the associated spatial barcodes allow the resulting sequencing reads to be assigned to their original capture locations. The data are then processed into gene-expression matrices linked to spatial coordinates. This approach provides broad gene-expression information across tissue sections, although the spatial resolution depends on the size and arrangement of the capture regions.
  • The original Spatial Transcriptomics technology, developed as an early foundation for modern spatial gene-expression analysis, introduced the use of spatially barcoded capture probes to connect RNA sequencing with tissue location. The method demonstrated that transcriptome-wide expression patterns could be mapped across intact tissue sections. It established an important experimental framework for later commercial and research platforms. Its development contributed to the wider adoption of spatially resolved molecular analysis and helped establish spatial transcriptomics as a major field within genomics.
  • 10x Genomics Visium is one of the widely used sequencing-based spatial transcriptomics platforms. It uses slides containing spatially barcoded capture spots, which capture RNA from tissue sections. The captured molecules are converted into sequencing libraries and analysed using next-generation sequencing. Each spot has a known position, allowing gene-expression measurements to be visualized on the corresponding tissue image. Traditional Visium measurements are generally spot-based rather than single-cell measurements, and individual spots may contain RNA from several neighbouring cells. Consequently, computational deconvolution may be required to estimate the cellular composition of each spot. Visium is useful for transcriptome-wide tissue mapping, identifying spatially variable genes, and comparing molecular patterns across anatomical regions.
  • Visium HD extends the Visium approach by providing much smaller spatial capture areas and higher-resolution gene-expression measurements. Its dense capture grid improves the ability to resolve molecular patterns at finer spatial scales. However, the resulting measurements may still require computational cell segmentation and assignment to individual cells because capture locations and actual cell boundaries are not necessarily identical. Visium HD is useful when researchers require broad gene-expression coverage combined with more detailed spatial localization than conventional spot-based approaches.
  • Slide-seq and Slide-seqV2 are sequencing-based methods that use arrays of spatially indexed beads to capture RNA molecules from tissue sections. Each bead carries a unique spatial barcode, allowing captured transcripts to be assigned to the bead’s known position. Slide-seq provides high spatial resolution compared with many traditional spot-based platforms, while Slide-seqV2 improves aspects of transcript detection sensitivity. These methods can reveal detailed spatial gene-expression patterns and help investigate the organization of cells within complex tissue structures. Their use requires specialized experimental procedures and computational processing of bead-level expression measurements.
  • HDST (High-Definition Spatial Transcriptomics) is another bead-based approach designed to achieve high spatial resolution through densely arranged spatially barcoded beads. It captures RNA molecules from small tissue locations, enabling detailed mapping of gene expression. Because individual capture locations may not correspond exactly to single cells, computational methods are needed to interpret the molecular composition of the measurements. HDST illustrates how increasing the density of spatial capture locations can improve the resolution of tissue molecular maps.
  • Open-ST is an open-source spatial transcriptomics approach based on spatially barcoded capture surfaces. It aims to provide an accessible framework for spatial RNA sequencing and can be adapted for different research applications. Like other capture-based technologies, it links RNA molecules to known positions through spatial identifiers and sequencing. Open-source approaches can be valuable for laboratories seeking greater experimental flexibility or alternatives to proprietary commercial systems, although implementation and performance depend on the specific protocol and laboratory setup.
  • Imaging-based spatial transcriptomics technologies detect RNA molecules directly within tissue sections. Rather than capturing all RNA molecules for conventional sequencing, these methods use fluorescent probes or other molecular recognition systems to identify selected transcripts. Repeated imaging cycles can increase the number of genes detected in the same tissue. The spatial coordinates of each detected molecule are recorded, allowing researchers to reconstruct gene-expression patterns at cellular or subcellular resolution. Imaging-based approaches are particularly useful for investigating rare cell populations, tissue boundaries, cellular neighbourhoods, and the localization of specific transcripts.
  • MERFISH (Multiplexed Error-Robust Fluorescence In Situ Hybridization) is an imaging-based technology that uses sets of fluorescently labelled probes and sequential imaging to identify multiple RNA molecules within individual cells. Each RNA target is assigned a molecular barcode represented by a pattern of fluorescent signals across imaging rounds. Error-robust coding helps distinguish targets despite missing or incorrect signal detection. MERFISH can provide high-resolution information about RNA localization and cellular organization. It is particularly useful for studying complex tissues in which the position of individual transcripts is biologically important.
  • seqFISH and seqFISH+ are sequential fluorescence in situ hybridization methods that use repeated rounds of hybridization and imaging to detect many RNA targets in tissue. Different combinations of fluorescent probes across imaging cycles create distinguishable molecular codes. seqFISH+ extends the approach to substantially larger gene panels. These methods can measure gene expression at cellular or subcellular resolution and are valuable for studying spatial gene-expression patterns, tissue organization, and cell-type distributions. Their performance depends on probe design, imaging quality, signal detection, and the number of genes included in the experiment.
  • 10x Genomics Xenium is a high-resolution, imaging-based spatial gene-expression platform that detects selected RNA targets directly in tissue sections. It uses probe-based molecular detection and imaging to identify transcripts and assign them to cellular locations. The platform supports cell segmentation and spatial visualization, enabling researchers to examine gene expression at cellular and subcellular scales. Xenium is particularly useful for targeted investigation of specific biological pathways, cell populations, and tissue regions. Its gene panel design means that the genes measured are generally selected in advance rather than obtained through unrestricted whole-transcriptome sequencing.
  • NanoString CosMx Spatial Molecular Imager is an imaging-based platform designed to measure RNA and, in relevant workflows, protein targets within tissue sections. It uses molecular probes and sequential imaging to identify selected targets and associate them with individual cells and spatial locations. CosMx can support high-plex molecular imaging and detailed analysis of cellular neighbourhoods. It is useful for investigating tumour microenvironments, immune-cell interactions, tissue organization, and molecular differences between neighbouring cell populations. As with other targeted imaging systems, the selected molecular panel influences which biological processes can be investigated.
  • In situ sequencing methods identify RNA sequences directly within tissue environments rather than relying exclusively on capture and sequencing of extracted RNA. STARmap (Spatially Resolved Transcript Amplicon Readout Mapping) is an example that combines molecular amplification and in situ detection to map gene expression at high spatial resolution. Other approaches, including FISSEQ (Fluorescent In Situ Sequencing), have explored direct sequencing of RNA-derived molecules within cells and tissues. These technologies provide opportunities for molecular analysis while retaining cellular architecture, although experimental complexity, sensitivity, and transcript coverage can differ considerably between methods.
  • Targeted in situ hybridization technologies also contribute to spatial transcriptomic analysis. RNAscope uses specialized probe designs and signal amplification to detect selected RNA molecules in tissue sections. It is widely used to validate gene-expression patterns identified through sequencing-based or other transcriptomic methods. RNAscope is generally a targeted detection method rather than a comprehensive transcriptome-wide spatial sequencing platform. It is valuable for confirming the localization of particular genes, identifying rare expressing cells, and validating spatial expression patterns using microscopy.
  • The major spatial transcriptomics platforms differ in spatial resolution, gene coverage, detection sensitivity, throughput, and experimental requirements. Traditional Visium provides broad transcriptome coverage across spatial spots, while Visium HD, Slide-seq, and HDST provide finer spatial capture. MERFISH, seqFISH, Xenium, and CosMx offer imaging-based molecular detection at cellular or subcellular resolution, usually using selected gene panels. In situ sequencing approaches provide another route to spatially resolved molecular identification. No single platform is ideal for every research question. A study focused on discovering broad tissue-expression patterns may benefit from transcriptome-wide capture, while a study examining specific cell interactions may require high-resolution imaging of selected genes.
  • Computational processing is essential for converting spatial transcriptomics measurements into biologically interpretable maps. Sequencing-based platforms generally produce raw sequencing reads that must be quality controlled, aligned or assigned to genes, and associated with spatial barcodes. Imaging-based platforms generate images and molecular detection signals that must be processed to identify transcripts and assign them to cells. Platform-specific software includes Space Ranger for 10x Genomics Visium data, Xenium Ranger for Xenium datasets, and other technology-specific processing pipelines. Quality control evaluates sequencing depth, detected genes, transcript counts, background signal, tissue coverage, segmentation quality, and the reliability of spatial assignments.
  • After initial processing, spatial gene-expression data can be analysed using general bioinformatics frameworks. Seurat supports spatial data visualization, clustering, normalization, and integration with single-cell datasets. Scanpy provides scalable single-cell analysis functions and can be used with spatial datasets through its ecosystem. Squidpy supports spatial statistics, spatial graphs, neighbourhood analysis, and tissue-level visualization. Giotto provides tools for spatial transcriptomics analysis, including spatial clustering and cellular interaction studies. Additional tools include SpatialExperiment, STUtility, and SPATA2. The appropriate software depends on the platform, data format, analytical objectives, and computational resources.
  • An important computational challenge is identifying the cell types contributing to spatial measurements. In spot-based methods, each measurement may contain RNA from several cells, making direct cell-type identification difficult. Spatial deconvolution methods estimate the proportions of cell types within each spot using reference expression profiles from single-cell or single-nucleus RNA sequencing. Important methods include RCTD, cell2location, SPOTlight, stereoscope, DestVI, and Tangram. Their outputs can be used to generate cell-type distribution maps and investigate differences in tissue composition. Imaging-based technologies may provide individual-cell measurements, but they still require accurate cell segmentation and annotation.
  • Spatial transcriptomics also enables the identification of spatially variable genes, which are genes whose expression changes systematically across tissue locations. These genes may identify tissue boundaries, developmental regions, disease-associated structures, and specialized cellular environments. Methods such as SpatialDE, SPARK, and SPARK-X can identify spatial patterns in gene-expression data. Spatial neighbourhood analysis can further investigate how different cell types are organized relative to one another. Tools such as Squidpy and Giotto support the construction of spatial graphs and analysis of cellular neighbourhoods. These methods can be combined with ligand–receptor analysis to explore potential cell–cell communication within tissues.
  • Integration with single-cell RNA sequencing is another major application. Single-cell data provide detailed gene-expression profiles and can serve as references for identifying cell types in spatial datasets. Spatial transcriptomics then provides information about where those cell types are located within the tissue. Integration methods can also help investigate cell states, tissue-specific expression, and molecular differences between spatial regions. Combining spatial transcriptomics with spatial proteomics, chromatin accessibility, and metabolomics creates opportunities for spatial multi-omics analysis, connecting gene expression with protein abundance, regulatory activity, and metabolic function.
  • Spatial transcriptomics technologies have extensive applications in cancer research, developmental biology, neuroscience, immunology, and tissue physiology. In cancer, they can reveal tumour heterogeneity, immune infiltration, stromal organization, vascular structures, and spatially restricted gene-expression patterns. In developmental biology, they can identify molecular gradients and cell populations involved in tissue and organ formation. In neuroscience, they help characterize the molecular organization of brain regions and neuronal populations. In immunology, they can reveal the distribution of immune cells and local inflammatory signalling. They are also valuable for investigating tissue repair, regeneration, fibrosis, infection, and metabolic organization.
  • In animal genomics and breeding research, spatial transcriptomics technologies can help investigate tissue-specific mechanisms underlying economically important traits. In livestock species such as cattle, sheep, goats, pigs, and poultry, spatial gene-expression mapping can reveal the organization of muscle, mammary, reproductive, immune, and metabolic tissues. In skeletal muscle, it can help identify the spatial distribution of muscle fibres, satellite cells, fibroblasts, endothelial cells, and immune cells involved in growth and regeneration. In mammary tissue, it can reveal the organization of epithelial and stromal populations associated with mammary development and milk production. In reproductive organs, it can support the investigation of follicular development, spermatogenesis, and embryo development. Comparative spatial analysis between breeds and developmental stages may identify molecular mechanisms associated with growth, fertility, disease resistance, and environmental adaptation.
  • Despite their advantages, spatial transcriptomics technologies have several limitations. Sequencing-based methods may combine transcripts from multiple cells within a capture spot, while imaging-based methods may be limited by the number of genes included in the panel. Detection sensitivity varies between technologies, and low-abundance transcripts may be missed. Tissue preservation, sectioning, RNA degradation, probe performance, imaging quality, and segmentation accuracy can influence data quality. Different platforms also generate different data structures, making cross-platform comparison and integration challenging. Spatial datasets may be computationally demanding and require substantial storage, processing resources, and specialist expertise.
  • The interpretation of spatial transcriptomics results also requires caution. RNA abundance does not always correspond to protein abundance or functional activity. A gene expressed in a particular region does not necessarily indicate that its encoded protein is active, and spatial proximity between two cell populations does not establish direct communication. Biological replication and validation are therefore essential. Histology, immunohistochemistry, immunofluorescence, RNAscope, protein assays, and functional experiments can provide independent evidence for spatial molecular findings. Combining complementary technologies often provides more reliable conclusions than relying on one platform alone.
  • Overall, spatial transcriptomics technologies have transformed the study of gene expression by preserving the relationship between molecular information and tissue architecture. Sequencing-based, imaging-based, and in situ sequencing approaches provide different ways to investigate molecular activity within intact tissues. Their integration with single-cell RNA sequencing, spatial proteomics, computational deconvolution, and spatial network analysis offers increasingly detailed views of cellular organization and biological function. Continued advances in resolution, sensitivity, transcript coverage, and computational analysis will strengthen their applications in medicine, developmental biology, veterinary science, comparative genomics, and animal genetic improvement.
Author:

Leave a Reply

Your email address will not be published. Required fields are marked *