- Biological systems are not organized only by genes, proteins, metabolites and signaling pathways; they are also organized in space. A molecule can have very different biological effects depending on where it is produced, which cells are nearby, and how it is distributed within a tissue. Spatial biology and spatial omics address this principle by measuring molecular features while preserving information about their physical location within cells, tissues or organs. Instead of asking only which genes, proteins or metabolites are present, spatial approaches ask where they are located, which cells contain them, which neighboring cells interact with them, and how molecular states are arranged across tissue architecture.
- This spatial perspective extends the concepts introduced by single-cell analysis and multi-omics integration. Single-cell methods can reveal molecular differences between individual cells, while spatial methods add information about the position of those cells within their tissue environment. A cell expressing a particular cytokine, receptor, transcription factor or metabolic enzyme may have a very different biological role depending on whether it is located next to an epithelial barrier, a blood vessel, an immune-cell cluster, a tumor region or a damaged area. Spatial information therefore provides an important connection between molecular measurements and tissue-level organization.
- Spatial transcriptomics is one of the most widely recognized forms of spatial omics. It measures RNA transcripts while retaining their spatial coordinates in a tissue section. Depending on the technology, spatial transcriptomics may measure transcripts across relatively large tissue spots or use imaging-based approaches capable of resolving individual cells or even subcellular RNA localization. This makes it possible to construct spatial maps of gene expression, identify transcriptionally distinct regions, and investigate how gene expression changes across tissue structures. Spatial transcriptomics is particularly useful when combined with single-cell RNA sequencing, because single-cell datasets can provide detailed molecular cell-type references that help interpret spatial expression patterns.
- Spatial proteomics extends the same principle to proteins. Because proteins are the functional molecules that directly perform many cellular processes, their spatial distribution can provide information that cannot always be inferred from RNA measurements. Spatial proteomics can identify proteins within individual cells, cellular compartments or tissue regions and can reveal the organization of signaling proteins, receptors, structural proteins, immune markers and other molecular components. Imaging-based methods can sometimes determine the localization of many proteins simultaneously, allowing researchers to investigate complex cellular neighborhoods while preserving tissue architecture.
- Spatial analysis can also be applied to epigenomics and chromatin biology. Spatially resolved measurements of chromatin accessibility, DNA methylation or other regulatory features can help connect tissue location with gene regulation. This is important because cells occupying different regions of the same tissue can have different regulatory states even when they share the same genome. Spatial epigenomics therefore helps explain how environmental signals, developmental programs and cellular interactions influence gene activity in specific tissue locations.
- Spatial metabolomics adds another layer by measuring metabolites and other small molecules in their tissue context. Metabolites are often highly sensitive to local physiological conditions, including oxygen availability, nutrient supply, cellular activity and metabolic interactions between neighboring cells. Spatial metabolomics can therefore reveal metabolic gradients and localized biochemical environments that may be hidden when metabolites are measured from an entire tissue homogenate. Similar spatial approaches can be extended to lipidomics, glycomics and other molecular layers.
- Different spatial technologies use different experimental principles. Sequencing-based spatial methods generally associate molecular measurements with predefined spatial coordinates, whereas imaging-based methods detect molecules directly within tissue using microscopy, fluorescence, probes, antibodies or other molecular recognition strategies. Some approaches prioritize broad molecular coverage, while others provide very high spatial resolution. The choice of method therefore involves a trade-off between resolution, sensitivity, molecular breadth, tissue compatibility, throughput and experimental complexity.
- One of the major advantages of spatial biology is its ability to preserve tissue architecture. Tissues are organized structures rather than random collections of cells. Epithelial layers, blood vessels, connective tissue, neuronal structures, immune-cell compartments and tumor regions have characteristic spatial arrangements. Spatial omics can associate molecular states with these structures and reveal how molecular programs correspond to anatomical organization. This allows researchers to investigate biological processes simultaneously at the molecular, cellular and tissue levels.
- Spatial data are particularly powerful for studying cellular neighborhoods. A cellular neighborhood is a local region containing particular combinations of cell types and molecular states. Two cells that belong to the same broad cell type may behave differently because they occupy different neighborhoods. For example, immune cells located close to tumor cells may experience different signaling environments from immune cells located near blood vessels or normal tissue. Spatial analysis can therefore reveal functional states that may not be obvious from cell identity alone.
- This spatial organization is closely related to cell-cell communication. Cells communicate through secreted molecules, membrane-bound ligands and receptors, extracellular matrix interactions and direct physical contacts. Spatial information makes it possible to ask whether potential ligand-producing and receptor-expressing cells are actually located close enough for a proposed interaction to be biologically plausible. Ligand-receptor analysis can therefore be strengthened by spatial information, although physical proximity alone does not prove that a signaling interaction is occurring.
- Spatial biology is especially important in cancer research. Tumors contain malignant cells together with immune cells, fibroblasts, endothelial cells, pericytes and other components of the tumor microenvironment. These populations can form distinct spatial regions and interact through signaling molecules, extracellular matrix components and metabolic pathways. Spatial omics can identify tumor borders, immune-excluded regions, invasive fronts, vascular niches and other molecularly distinct areas. It can also reveal differences between tumor cells located in different microenvironments, helping researchers understand tumor heterogeneity beyond differences in genetic or transcriptional state alone.
- The same principles apply to developmental biology. During development, cells change their identity while moving through a spatially organized embryo or developing organ. Spatial transcriptomics and related methods can map gene-expression programs onto anatomical structures and reveal how developmental gradients and cellular neighborhoods emerge. When combined with cell lineage analysis, trajectory analysis and single-cell datasets, spatial information can help connect molecular cell states with their developmental positions.
- Spatial biology is also highly valuable in neuroscience, where physical organization is fundamental to function. Neurons, glial cells, blood vessels and extracellular structures occupy highly organized locations within the nervous system. Spatial molecular maps can associate gene expression, protein expression and cellular states with anatomical regions and neuronal circuits. This provides an important connection between molecular neuroscience, cellular organization and tissue architecture.
- In immunology, spatial methods can reveal how immune cells are distributed within tissues and how they interact with infected, damaged or transformed cells. The organization of immune-cell populations within lymphoid organs, tumors and inflamed tissues can influence immune activation and suppression. Spatial measurements can therefore complement conventional immunophenotyping by showing not only which immune populations are present but also where they are located and which molecular environments surround them.
- A central computational challenge in spatial omics is cell-type identification. Some spatial technologies measure molecules at approximately cellular resolution, while others measure groups of neighboring cells within spatial spots. When individual cells are not completely resolved, computational deconvolution can estimate which cell types contribute to each spatial measurement. Single-cell RNA-sequencing datasets can serve as reference maps for this process. The integration of single-cell and spatial datasets therefore allows researchers to combine the molecular resolution of single-cell sequencing with the anatomical information provided by spatial measurements.
- Spatial datasets can also be analyzed using dimensionality reduction, clustering, spatial statistics and machine learning. Conventional clustering can identify molecularly distinct regions, while spatially aware methods incorporate the coordinates of measurements into the analysis. Graph-based analysis can represent cells or spatial locations as nodes connected according to physical proximity or molecular relationships. This makes it possible to identify spatial communities, interaction networks and tissue modules.
- The concept of a spatially organized network extends naturally from network biology and systems biology. Instead of representing a biological network only as molecular interactions, researchers can incorporate spatial coordinates into the network. A signaling network, for example, can be analyzed together with the locations of the cells producing and receiving signals. Similarly, metabolic pathways can be connected to spatially localized metabolite concentrations, while gene regulatory networks can be related to the anatomical regions in which particular regulatory programs are active.
- Spatial omics can therefore be integrated with multi-omics data integration. Genomics can provide information about genetic variation, transcriptomics can reveal gene-expression states, epigenomics can describe regulatory states, proteomics can measure proteins, and metabolomics can describe biochemical activity. Spatial measurements add the location of these molecular states. Together, these datasets can create a multidimensional representation of tissue organization in which genotype, regulation, expression, protein activity, metabolism and spatial position are analyzed as interconnected properties.
- The integration of spatial and single-cell datasets is particularly important because the two approaches provide complementary information. A single-cell dataset may provide extremely detailed molecular profiles but lose the original location of each cell. A spatial dataset preserves location but may provide fewer measured genes or proteins per cell. Computational integration can use one dataset as a molecular reference and the other as a spatial framework, allowing researchers to infer detailed cell states across tissue architecture.
- Spatial information also becomes important when studying genetic variants. A mutation may occur throughout a tissue but produce its biological consequences only in particular cell types or anatomical regions. Integrating genomic variation with spatial transcriptomics and proteomics can help connect genotype with local molecular phenotypes. For example, a variant affecting a regulatory element may alter gene expression specifically in a particular cellular population or tissue compartment. This creates a bridge between human genetics, variant interpretation, molecular mechanisms and tissue pathology.
- Spatial approaches can also be connected with protein structure and structural bioinformatics. A spatially localized protein can be analyzed at the molecular level using information about its domains, motifs, three-dimensional structure, binding sites and interactions. If a mutation changes a protein’s structure or ligand-binding properties, spatial measurements can help determine where cells expressing that altered protein are located and which neighboring cells may be affected. Thus, molecular structure and tissue organization represent different but complementary spatial scales.
- The same principle applies to drug discovery. A drug does not act on an isolated molecular target in an abstract network. It reaches particular tissues, encounters specific cell types, interacts with proteins and may produce different responses in different spatial environments. Spatial omics can be used to study drug response, target expression, pathway activation and treatment-associated changes across tissue regions. In cancer research, for example, spatial measurements can help identify tumor regions that respond differently to treatment and may reveal localized molecular environments associated with drug resistance.
- Spatial data can also contribute to understanding drug-target interactions and polypharmacology. If several drug targets are expressed in different cellular populations, their spatial distribution may help explain why a drug produces effects in one tissue compartment but not another. Similarly, off-target effects may depend not only on molecular affinity but also on target expression, cellular identity, tissue distribution and drug exposure. Spatial biology therefore provides an additional layer between molecular drug-target interactions and whole-tissue pharmacological responses.
- Spatially resolved pathway analysis is another important application. Instead of asking whether a pathway is activated somewhere in a tissue, researchers can determine where its activity-associated molecular signatures are concentrated. Signaling pathways, metabolic pathways, inflammatory programs, developmental pathways and stress responses can exhibit strong spatial variation. Mapping pathway activity onto tissue architecture can reveal molecular gradients, boundaries and localized biological states.
- Spatial biology can also incorporate time. Traditional spatial measurements often provide a snapshot of tissue organization, but biological processes are dynamic. Combining spatial measurements from different time points can reveal how molecular patterns move or change during development, disease progression, tissue repair or drug treatment. When combined with time-series analysis, perturbation experiments and single-cell trajectories, spatial data can help distinguish transient molecular states from stable tissue compartments.
- One emerging direction is spatial single-cell multi-omics, in which multiple molecular layers are measured while preserving cellular location. Such experiments can potentially connect gene expression, chromatin accessibility, proteins and other molecular features within the same spatial framework. These approaches are technically challenging because each additional molecular layer introduces requirements for sensitivity, molecular specificity, imaging, sequencing and computational integration. Nevertheless, they provide a route toward highly detailed maps of tissue organization.
- Spatial omics also requires careful consideration of experimental quality and computational uncertainty. Tissue preparation can alter RNA, proteins and metabolites. Fixation and processing may affect molecular detection, while sectioning can introduce distortions. Limited sensitivity can result in missing molecular measurements, and imaging-based approaches can encounter segmentation errors. Sequencing-based methods may have limited spatial resolution, while high-resolution methods may measure fewer molecular species. Batch effects, differences between tissue samples and computational normalization can further influence the final interpretation.
- Another challenge is distinguishing spatial association from biological causation. If two cell types are located next to each other and express complementary signaling molecules, this provides evidence consistent with possible communication but does not prove that the interaction occurs. Functional experiments, perturbation studies, receptor activity measurements and other independent evidence may be required. Spatial biology therefore works most effectively when combined with molecular, cellular and experimental validation.
- A typical spatial omics workflow begins with tissue or organ selection and appropriate sample preservation, followed by spatial molecular measurement and imaging or sequencing. The resulting molecular measurements are associated with spatial coordinates and undergo quality control, normalization and computational processing. Cells or spatial regions can then be identified, annotated using single-cell or reference datasets, and analyzed for spatial patterns. Researchers may subsequently perform differential expression, pathway analysis, cell-cell communication analysis, spatial statistics and network analysis before integrating the findings with genomics, proteomics, metabolomics or other datasets. Experimental validation can then test important biological hypotheses.
- Spatial biology therefore adds another dimension to the molecular hierarchy developed throughout this series. Protein sequence analysis can identify sequence similarity, protein families and protein domains. Protein motifs and signatures can provide clues about molecular function, while protein domain architecture describes the organization of functional regions. Protein structure prediction, structural alignment and visualization can reveal three-dimensional organization, while protein-ligand interactions, molecular docking, molecular dynamics and binding free-energy calculations describe molecular interactions. These molecular mechanisms connect to drug-target interactions, polypharmacology, network pharmacology and systems biology. Multi-omics then connects multiple molecular layers, single-cell multi-omics resolves these layers between individual cells, and spatial omics adds the crucial information of where those cells and molecular processes occur within tissues.
- The resulting framework can be viewed as a progression from sequence to structure, interaction, pathway, network, cell, tissue and phenotype. Each level provides information that cannot be completely recovered from the previous level alone. Spatial biology is therefore not simply another omics technology; it is a way of connecting molecular measurements to the physical organization of living systems.
