Spatial transcriptomics allows researchers to capture gene expression while maintaining tissue architecture by combining traditional transcriptomics with spatial information. This method counts the number of gene transcripts at distinct spatial locations using in situ hybridization, sequencing, and capturing technologies, providing a detailed transcriptome map across tissue sections.

Unlike traditional transcriptional studies that analyze homogenized tissues, spatial transcriptomics retain the spatial context, offering a vivid picture of gene expression within its natural environment. This resolution uncovers tissue architecture, cellular interactions, and functional dynamics in specific microenvironments. It is sensitive enough to reveal relationships between genes and their immediate neighbors, insights previously obscured in bulk transcriptomic studies.

Spatial transcriptomics is transforming the study of cellular organization, interactions, and biological mechanisms, advancing our understanding of health and disease across fields such as cancer research, biomarker discovery, neuroscience, developmental biology, and immunology.

Key techniques and technologies in spatial transcriptomics

Spatial transcriptomics encompasses various techniques that can be broadly categorized into sequencing-based methods, imaging-based methods, and hybrid methods. Each category employs different strategies to capture spatial gene expression data.

Sequencing-based methods

Sequencing-based methods include the extraction of RNA from tissue sections, reverse-transcribing it into cDNA, and sequencing it to give a high-resolution gene expression map through the barcoding technique¹. These methods are particularly beneficial when high-throughput and large-scale gene expression data are needed, and they can provide a comprehensive analysis of gene activity across a tissue’s architecture. The biggest challenge for this method is maintaining the spatial identity of each transcript in the tissue.

In a study on glioblastoma, researchers used a spatial transcriptomics platform to map the spatial expression of genes within the tumor microenvironment². This allowed them to understand the distinct gene expression profiles in tumor cells versus surrounding healthy tissue, providing new insights into how the tumor interacts with its microenvironment and identifying potential therapeutic targets.

The cutting-edge spatial transcriptomics method was first introduced by Stahl and colleagues, who used sequencing-based spatial transcriptomics technology. This method enables researchers to map gene expression across tissue samples with high spatial resolution. The technology utilizes an array of spatially indexed barcoded spots (55 μm in diameter) placed on glass slides, which capture RNA molecules from the tissue. These spots are linked to specific locations on the tissue, allowing for precise mapping of gene expression patterns^3,4,5.

After tissue fixation and permeabilization, mRNA diffuses into the arrayed spots, where it is captured by oligo(dT) probes. Reverse transcription is then performed, and the cDNA is processed for sequencing, producing a detailed gene expression profile that is spatially mapped to its original location.

Slide-seq and slide-seqV2

Slide-seq uses a novel method for placing tissue pieces on a surface coated with DNA-barcoded beads3,6. The highly spotted barcoded DNA beads on a slide allow for the capture of RNA from adjacent tissue slices in slide-seq and its enhanced variant, slide-seqV2. It outperforms previous techniques in terms of spatial resolution and enables fine-scale RNA mapping at high density. The high-density bead array enables the detection of spatially resolved gene expression at a resolution of 10 μm, which is close to single-cell resolution. This makes slide-seqV2 significantly effective for capturing subtle details in tissue architecture and gene activity.

Slide-seqV2 is very sensitive to detecting low-abundance transcripts that are important for analyzing delicate biological processes.

In a study of the mouse hippocampus, slide-seqV2 was used to reveal detailed spatial patterns of gene expression related to memory and learning.  Here, slide-seqV2 detected 550 unique molecular identifiers (UMIs) per unit. This method also helped detect 45772 UMIs per 100 µm2. This fine-scale resolution helped identify subtle regional variations in gene expression that could be linked to cognitive dysfunction in neurodegenerative diseases^3,7.

Additionally, slide-seqV2’s high resolution has allowed for significant advances in developmental biology. It is specifically useful for mapping gene expression in early-stage embryos, where tissue is often small and densely packed, making it hard for other spatial transcriptomics methods to provide clear insights.

Deterministic barcoding in Tissue (DBIT)-seq

This method was developed by Rong Fan’s group in 2020 for detecting both mRNAs and proteins in formalin-fixed paraffin-embedded (FFPE) and frozen tissue sections. This technique uses a polydimethylsiloxane (PDMS) microfluidic chip with 50 parallel microchannels, each filled with a different barcoded oligo solution. The microfluidic chip captures mRNA through an oligo(dT) domain, followed by reverse transcription.

A second perpendicular chip with additional barcoded oligos is introduced, and T4 ligase facilitates the ligation of oligos, creating unique spatial barcodes that map to specific tissue regions. This process generates a spatially barcoded mosaic, enabling the identification of gene expression patterns in relation to tissue architecture, followed by cDNA collection for next-generation sequencing (NGS). DBIT-seq offers precise spatial resolution for transcriptomic and proteomic analyses, facilitating comprehensive tissue profiling^3,8.

Imaging-based methods

Imaging-based approaches directly visualize mRNA molecules in tissue sections, thus combining spatial specificity with molecular specificity. It is helpful for detecting gene-expression patterns at resolutions of the single-cell or subcellular types. They allow researchers to observe gene expression at the cellular or even subcellular level, providing rich information about the spatial organization of tissues and the relationships between cells. The primary advantage of imaging-based methods is their ability to reveal the exact location of gene expression within the tissue architecture.

Fluorescence in-situ hybridization (FISH)

FISH has proven valuable in clinical diagnostics, especially in cancer. For example, it has been used to detect fusion genes like BCR-ABL in leukemia, providing vital information about the spatial localization of gene rearrangements in tissue biopsies.

This is a well-established yet potent imaging technique for spatial transcriptomics. Fluorescently labeled probes hybridize specifically to target RNA sequences in tissue sections. Then, the probe captures the fluorescence signals, enabling researchers to obtain high spatial-resolution images of gene expression patterns. FISH is particularly advantageous when investigating the distribution of particular genes or transcripts in tissues across different cell types but is limited in multiplexing capabilities.

A study on Arabidopsis thaliana (a model plant) used FISH to investigate the expression of key genes (AP1, AP2, AP3) involved in floral organ development. By combining multiple fluorescent probes, they successfully visualized the spatial distribution of genes responsible for floral patterning, offering insight into the developmental processes⁹.

Multiplexed error robust FISH (MERFISH)

MERFISH is an advanced version of FISH. It facilitates the detection of multiple RNA species in one tissue section. With the help of error-robust coding schemes like the modified Hamming distance of 4 (MHD4), MERFISH possesses high sensitivity and specificity and enables one to profile hundreds to thousands of genes from a single sample. MERFISH works by encoding the target RNA sequences with a unique set of fluorescent signals, which helps to reduce background noise and improve the sensitivity of detection. This makes MERFISH ideal for studying complex gene expression patterns in tissues with heterogeneous populations of cells, such as the brain or tumors^1,10.

The unique bit system used in MERFISH involves encoding each gene with an N-bit binary code, which is detected through successive rounds of hybridization with encoding and readout probes. In this system, “bit-1” indicates the presence of a signal, while “bit-0” indicates the absence of a signal. The bit code is decoded after N rounds, with each round determining the presence or absence of specific sequences flanking the target gene.

In a study of the mouse brain, MERFISH was used to simultaneously profile the expression of hundreds of genes involved in neurodevelopment and disease. The technique enabled the identification of distinct molecular layers within the cortex, shedding light on the complex molecular architecture of the brain. MERFISH also helps to identify specific gene expression patterns linked to neurological disorders such as autism spectrum disorders¹¹.

Single-molecule FISH (smFISH)

This method involves fixing, permeabilizing, and hybridizing cells with fluorescently tagged DNA probes that tile the mRNA. By improving the signal-to-noise ratio, the probe multiplicity makes it possible to detect them using microscopy as diffraction-limited spots. These locations are found using a 3D Gaussian fitting method. smFISH maintains spatial RNA localization while detecting single to hundreds of RNA molecules. It provides accurate RNA quantification and localization information for a variety of RNA types, including mRNA, lncRNA, viral RNA genomes, and rRNA¹².

smFISH can visualize single RNA molecules inside cells. Due to this high sensitivity, it can quantify low-abundance transcripts with exceptional precision¹³. Such a tool is extremely valuable for the study of gene expression in rare or transient cellular states such as cell differentiation or response to stimulus.

In a study, researchers applied smFISH to visualize and quantify the distribution of mRNA transcripts within the biofilm¹⁴. By detecting single RNA molecules, they were able to track how gene expression varies within different regions of the biofilm, uncovering gene regulatory networks important for bacterial survival in hostile environments.

smFISH has been used to track the expression of stem cell markers in different developmental stages in Drosophila (fruit fly)¹⁵, which can allow researchers to understand how gene expression drives cellular differentiation during development.

Key factors for selecting spatial transcriptomic techniques

The choice of the optimal spatial transcriptomic technique relies on several considerations, ranging from gene throughput, sequence information, and sensitivity to resolution, area size, and feasibility⁴.

Gene throughput: Methods based on next-generation sequencing (NGS) are unbiased since they capture every polyadenylated transcript without prior knowledge of gene targets and are thus particularly suitable for characterizing new systems. Nonetheless, in situ hybridization (ISH) and the majority of in situ sequencing (ISS) techniques, which are typically targeted, are based on the initial identification of genes of interest. Notwithstanding this, the throughput of targeted approaches has improved enormously in recent years to as much as 10,000 genes. Targeted approaches can be combined with single-cell RNA-Seq, allowing for more accurate localization of previously identified genes.

Sequence data: Both ISS and NGS-based techniques provide deep sequence information, allowing for the identification of splice isoforms, single nucleotide variants, and point mutations. This can be useful in building gene expression time courses or lineage tracing, particularly when coupled with RNA velocity.

Sensitivity: ISH-based approaches are extremely sensitive, with a detection efficiency of up to 80%, near that of the gold standard single-molecule FISH (smFISH). Although NGS-based approaches are less sensitive and presently worse than single-cell RNA-Seq, their sensitivity is increasing very fast, reaching up to 100 unique transcripts per square micrometer. There is generally a compromise between sensitivity and gene throughput, with targeted ISS approaches being more sensitive than unbiased approaches.

Resolution: ISH-based techniques, particularly when combined with methods such as expansion microscopy, provide high resolution, as low as 100 nm, and are well suited for examining sub-cellular organization. NGS-based techniques are less resolved but have also improved considerably, with current techniques providing a resolution of around 1 μm.

Area size: In situ procedures have the advantage of flexibility in sample size, but imaging conditions take time as the tissue section is larger. NGS-based procedures rely on standard arrays with small area coverage, which could be challenging for extremely small or extremely large samples.

Feasibility: The availability of some of these high-tech approaches, eg, single-molecule imaging for ISH or customized capture arrays, may be restricted. Though commercialization has facilitated greater availability, these technologies may still have challenges in cost and infrastructure.

Selection of an appropriate method is about weighing these issues against each other depending on the goals of the research and the available resources.

Data analysis and bioinformatics for spatial transcriptomics

Data generated by spatial transcriptomics can only be handled using sophisticated computational tools and bioinformatics methodologies. These tools are essential for managing spatially resolved gene expression data, which integrates molecular information with spatial context, enabling the discovery of spatially distinct cell populations and their interactions within tissues. In this respect, there are various bioinformatics platforms tailored to the unique challenges spatial transcriptomics data poses.

Data analysis tools and methodologies

Key steps in analyzing spatial transcriptomics data include^16,17:

Normalization: Adjusting raw data to account for technical variability.

Dimensionality reduction: Techniques like principal component analysis (PCA) or uniform manifold approximation and projection (UMAP) are used to reduce data complexity while preserving meaningful patterns.

Clustering: Identifying groups of cells with similar expression profiles to define cell types or states.

Visualization: Using spatial heatmaps or 3D reconstructions to overlay gene expression patterns onto tissue architecture.

For example, some tools, such as the ST pipeline and slideseq-tools process raw spatial transcriptomics data into count matrices, aligning tissue images with gene expression profiles¹⁶. Similarly, some tools employ Bayesian statistical models such as BayesSpace, to infer latent spatial domains, offering insights into complex tissue structures18.

Computational challenges and solutions

Spatial transcriptomics has recently made advances, although the technique faces challenges with computational problems^16,17.

High dimensionality: The datasets are large and multidimensional, requiring efficient algorithms for processing.
Spatial interdependence: Gene expression is influenced by spatial context, necessitating models that account for spatial correlations.

Data integration: Combining spatial transcriptomics with single-cell RNA sequencing (scRNA-seq) adds complexity due to differences in resolution and experimental conditions.

Various methods can be used to address these challenges^16,17:

• Methods like SpatialDE and SpaGE integrate gene expression with spatial coordinates to identify spatially coherent patterns.
• Deep learning models are increasingly used for tasks like cell segmentation and prediction of unmeasured gene expressions. For example, some models employ generalized linear spatial models to predict missing gene profiles by integrating scRNA-seq data with spatial transcriptomics.
• Combining imaging-based methods (eg, MERFISH) with sequencing-based approaches enhances both resolution and coverage.

Applications of spatial transcriptomics

Spatial transcriptomics has numerous applications across several biological research fields, explaining gene expression patterns in different contexts.

Cancer research

Spatial transcriptomics provides detailed views of the cancer tumor microenvironment¹⁹. Using this approach, scientists can assess gene expression across different cell populations within a tumor and their interactions with the surrounding stromal and immune cells. Spatial mapping of transcriptomics aids in identifying therapeutic targets and enhances the understanding of how tumors develop, metastasize, and acquire resistance.

Mapping tumor heterogeneity: By spatially resolving gene expression, spatial transcriptomics can delineate distinct cell populations within tumors²⁰. For example, in breast cancer studies, researchers identified unique spatial domains of immune infiltration and hypoxia, which were associated with patient outcomes.

Therapeutic target identification: This technology has been used to uncover potential drug targets by mapping ligand-receptor interactions in the tumor microenvironment (TME). In lung cancer samples, spatial transcriptomics revealed interactions between cancer-associated fibroblasts and tumor cells that could be targeted for therapy²⁰.

Drug resistance mechanisms: Spatial transcriptomics has been applied to study how tumors acquire resistance to treatments. For example, it identified spatially confined subpopulations of resistant cells in melanoma treated with targeted therapies¹⁹.
Spatial microstructural biomarkers: Spatial transcriptomics technologies have also facilitated the identification of microstructural biomarkers that provide essential insights into tumor prognosis and treatment response. The 6-gene signature for hepatocellular carcinoma patient survival prediction and the GATA3 mutation for breast cancer relapse are two prominent examples¹⁰.

Neuroscience

Spatial transcriptomics plays a vital role in neuroscience by revealing the cellular composition and functional states of brain regions. Mapping gene expression in specific populations of neurons helps researchers understand the molecular basis of brain function, neurodevelopmental disorders, and neurodegenerative diseases. This ability to study gene expression also enhances our understanding of how different parts of the brain contribute to behavior and cognition.

Mapping brain regions

Researchers have used spatial transcriptomics to create detailed atlases of the mouse and human brain. For example, a multimodal atlas of the primary motor cortex revealed distinct neuronal subtypes and their spatial organization²¹.

Neurodegenerative diseases: In Alzheimer’s disease research, spatial transcriptomics has been used to identify microglial activation states and their distribution in diseased brain regions. This helps in understanding how inflammation contributes to neurodegeneration²².

Developmental neuroscience: Spatial transcriptomics has been applied to study brain development by mapping gene expression changes during cortical layer formation in embryos²³.

Developmental biology

Spatial transcriptomics is a powerful tool for understanding gene expression in the context of development. Directly comparing gene expression profiles at different stages of development can provide insights into the mechanisms involved in organogenesis, tissue differentiation, and morphogenesis. Additionally, it is highly valuable in studying model organisms to identify molecular signals that regulate tissue patterning and development.

Reconstructing developmental pathways: By combining single-cell RNA sequencing with spatial data, researchers can trace cell lineage trajectories. For example, in mouse embryonic liver studies, spatial transcriptomics revealed erythro-myeloid progenitor differentiation pathways²⁴.

Tissue patterning: In zebrafish embryos, spatial transcriptomics are used to identify molecular gradients that regulate tissue patterning during early development²⁵.

Stem cell dynamics: The technology has also been employed to study stem cell niches within developing tissues, providing insights into how stem cells contribute to organ formation.

Immunology

Spatial transcriptomics helps understand the spatial distribution and function of immune cells in tissues. Such information enables detailed mapping of immune responses that can be related to the localization of immune cell populations in infections or autoimmune diseases. Such spatial information is required to understand how immune cells interact with other cell types in disease contexts and to develop more targeted immunotherapies.

Immune cell profiling: The technology enables detailed mapping of immune cell subsets (eg, T cells and B cells) within tissues²⁶. For example, in melanoma studies, spatial transcriptomics identified immune-excluded regions where T cells were unable to infiltrate tumors.

Infection studies: Researchers have used this approach to study how immune cells respond to infections by mapping cytokine-producing cells in infected tissues²⁷.

Autoimmune diseases: In rheumatoid arthritis studies, spatial transcriptomics revealed the spatial distribution of inflammatory cytokines in synovial tissues, aiding in understanding disease mechanisms²⁸.

Immunotherapy insights: By profiling immune cells within the TME during checkpoint inhibitor therapy, researchers can identify responders versus non-responders based on spatial patterns of immune activation²⁹.

Imaging modality constraints: Different imaging techniques (eg, wide-field fluorescence microscopy, confocal microscopy, light sheet microscopy) have trade-offs between speed, depth resolution, and cost³⁰. For example, while light sheet microscopy offers superior optical sectioning for 3D samples, it requires significant investment and expertise.

Multifaceted datasets: Spatial transcriptomics generates highly complex datasets that integrate spatial coordinates, gene expression levels, and histological features. Analyzing these datasets requires sophisticated computational tools capable of managing spatial interdependence and high dimensionality.

Integrative approaches combining spatial transcriptomics with other modalities (eg, proteomics or imaging) further increase complexity but are essential for gaining deeper biological insights³¹.

Computational challenges: The large size and multidimensional nature of spatial transcriptomics data imposes significant computational burdens³². Advanced algorithms such as graph convolutional networks have been developed to improve spatial domain identification but require substantial computational resources.

Deep learning models are increasingly applied to analyze these datasets but add another layer of complexity due to the need for large training datasets and specialized hardware.

High costs: The cost of spatial transcriptomics experiments remains expensive for many research organizations^30,33. Efforts are underway to reduce costs through innovations in barcoding strategies and sequencing efficiency while maintaining resolution and sensitivity.

Limited accessibility: The high cost and technical expertise required limit accessibility to well-funded institutions. This creates disparities in the adoption of spatial transcriptomics across different regions and research fields.

Reproducibility: Variability in sample preparation protocols (eg, tissue sectioning or expansion processes) can affect reproducibility³⁰. For instance, ExM protocols may lead to RNA loss or inconsistent expansion ratios, complicating data interpretation.

Plant-specific challenges: In plant research, cell walls and secondary metabolites pose unique challenges for spatial transcriptomics due to difficulties in cryosectioning and mRNA release³³. Optimizing methods for plant tissues remains a significant hurdle.

Throughput vs. resolution trade-off: Balancing throughput with resolution is a persistent challenge³¹. High-resolution methods often sacrifice throughput, making it difficult to scale experiments for large tissue samples or multiple conditions.

Trends and future directions in spatial transcriptomics

Future innovations in spatial transcriptomics are likely to focus on enhancing both resolution and throughput capabilities. New technologies may emerge that enable even finer mapping of gene expression at subcellular levels while maintaining high-throughput capabilities.

Multi-omics integration

Integrating spatial transcriptomics with other omics technologies (such as proteomics or metabolomics) will provide a more comprehensive understanding of biological systems. This multi-omics approach can reveal intricate relationships between different molecular layers within tissues.

Clinical applications

As spatial transcriptomics technology advances, its impact on clinical applications will grow significantly. This includes applications in diagnostics, personalized medicine, and real-time in vivo monitoring of disease progression or treatment responses. By enabling in situ mapping of gene expression, spatial transcriptomics offers unprecedented insights into disease biology, with the potential to transform clinical practice and improve patient outcomes.

Additional emerging applications

Machine learning integration: Spatial transcriptomics data is increasingly being used as input for machine learning models. For example, deep-learning approaches have been applied to integrate spatial gene expression data with histological images to predict tumor morphology.

Drug development: Spatial transcriptomics aids drug discovery by identifying tissue-specific drug targets. For instance, it has been used to map chemokine receptor expression in inflammatory diseases for targeted therapy development.
Tissue engineering: In regenerative medicine, spatial transcriptomics may help in designing biomimetic tissues by providing a blueprint of the native tissue architecture.

Challenges and limitations of spatial transcriptomics

Spatial transcriptomics has revolutionized the study of gene expression by preserving spatial information within tissues. However, despite its transformative potential, technology faces several challenges and limitations that need to be addressed for broader adoption and enhanced utility.

Resolution and sensitivity: Current spatial transcriptomics platforms often struggle with achieving single-cell or subcellular resolution³⁰. For example, some technologies have a spot size of 55 µm, which can encompass multiple cells, leading to mixed RNA signals from different cell types.

Detecting low-abundance transcripts remains challenging due to optical crowding and the limited sensitivity of imaging methods. Techniques like expansion microscopy (ExM) have been introduced to improve resolution by physically enlarging tissue samples, reducing crowding, and enabling the detection of more transcripts per tissue area.

Imaging-based methods such as MERFISH and ISS are limited by RNA degradation during sample preparation and optical degradation during imaging cycles, which restricts the number of genes that can be interrogated simultaneously.

RNA length and variant detection: Short RNA reads generated by some platforms can lead to misidentification of transcripts or failure to detect splicing variants and single-nucleotide polymorphisms (SNPs)³⁰. For example, sequencing-by-ligation methods often produce reads of only 30 bases, making it difficult to resolve complex transcriptomes.

Innovations like spatial isoform transcriptomics are addressing this issue by enabling the detection of RNA isoforms in spatial data. This technique involves combining spatial information with transcriptomic data, allowing the identification and characterization of RNA isoforms variants in distinct cellular environments, such as within tissues or cells³⁴.

FAQs

How does MERFISH compare to other spatial transcriptomics methods?

MERFISH is unique among spatial transcriptomics methodologies because it can achieve single-cell resolution and subcellular spatial precision, outperforming typical sequencing-based approaches. Although it is limited to investigating hundreds or thousands of genes, MERFISH compensates for this constraint by delivering high sensitivity and robust cell type identification without the requirement for intensive computational integration with other datasets.

What are the main applications of spatial transcriptomics in cancer research?

Spatial transcriptomics helps identify tumor heterogeneity, map tumor microenvironments, and understand immune cell infiltration and interactions. It also reveals gene expression patterns linked to tumor progression, resistance, and therapeutic targets.

How does spatial transcriptomics enhance our understanding of cellular interactions?

Spatial transcriptomics preserves tissue architecture, allowing researchers to study how cells interact within their natural environment. It uncovers relationships between genes and neighboring cells, providing insights into cellular communication, microenvironments, and functional dynamics.

References

1. Chen, T. Y., You, L., Hardillo, J. A. U., et al.  Spatial transcriptomic technologies. Cells, 12, 2042 (2023).
2. Shireman, J. M., Cheng, L., Goel, A., et al. Spatial transcriptomics in glioblastoma: is knowing the right zip code the key to the next therapeutic breakthrough? Frontiers in Oncology, 13, 1266397 (2023).
3. Robles-Remacho, A., Sanchez-Martin, R. M., & Diaz-Mochon, J. J. Spatial transcriptomics: emerging technologies in tissue gene expression profiling. Analytical Chemistry, 95, 15450-15460 (2023).
4. Rao, A., Barkley, D., França, G. S., et al. Exploring tissue architecture using spatial transcriptomics. Nature, 596, 211-220 (2021).
5. Tian, L., Chen, F., & Macosko, E. Z. Moving genomics into tissues. Nature biotechnology, 41, 773 (2022).
6. Rodriques, S. G., Stickels, R. R., Goeva, A., et al. Slide-seq: A scalable technology for measuring genome-wide expression at high spatial resolution. Science, 363, 1463-1467 (2019).
7. Stickels, R. R., Murray, E., Kumar, P., et al. Highly sensitive spatial transcriptomics at near-cellular resolution with Slide-seqV2. Nature biotechnology, 39, 313-319 (2021).
8. Liu, Y., Yang, M., Deng, Y., et al. High-spatial-resolution multi-omics sequencing via deterministic barcoding in tissue. Cell, 183, 1665-1681 (2020).
9. Yao J.L., Kang C., Gu C., Gleave A.P. The roles of floral organ genes in regulating Rosaceae fruit development. Front Plant Sci, (2022).
10. Wang, Y., Liu, B., Zhao, G., et al., Spatial transcriptomics: Technologies, applications and experimental considerations. Genomics, 115, 110671 (2023).
11. Fang R., Xia C., Close J.L., et al. Conservation and divergence of cortical cell organization in human and mouse revealed by MERFISH. Science, 377, 56-62 (2022).
12. Haimovich G., Gerst J.E. Single-molecule Fluorescence in situ Hybridization (smFISH) for RNA Detection in Adherent Animal Cells. Bio Protoc, 8, (2018).
13. Haimovich, G., & Gerst, J. E. Single-molecule fluorescence in situ hybridization (smFISH) for RNA detection in adherent animal cells. Bio-protocol, 8, (2018).
14. Barbosa, A., Miranda, S., Azevedo, et al., Imaging biofilms using fluorescence in situ hybridization: Seeing is believing. Frontiers in cellular and infection microbiology, 13, 1195803 (2023).
15. Trcek, T., Lionnet, T., Shroff, H., et al., mRNA quantification using single-molecule FISH in Drosophila embryos. nature protocols, 12, 1326-1348 (2017).
16. Liu, B., Li, Y., & Zhang, L. Analysis and visualization of spatial transcriptomic data. Frontiers in Genetics, 12, 785290 (2022).
17. Yue, L., Liu, F., Hu, J., et al. A guidebook of spatial transcriptomic technologies, data resources and analysis approaches. Computational and Structural Biotechnology Journal, 21, 940-955 (2023).
18. Zhao E., Stone M.R., Ren X., et al. Spatial transcriptomics at subspot resolution with BayesSpace. Nat Biotechnol, 39, 1375-1384 (2021).
19. Yu, Q., Jiang, M., & Wu, L. Spatial transcriptomics technology in cancer research. Frontiers in Oncology, 12, 1019111 (2022).
20. Cilento, M. A., Sweeney, C. J., & Butler, L. M. Spatial transcriptomics in cancer research and potential clinical impact: a narrative review. Journal of Cancer Research and Clinical Oncology, 150(6), 296 (2024).
21. Bhattacherjee, A., Zhang, C., Watson, B. R., et al. Spatial transcriptomics reveals the distinct organization of mouse prefrontal cortex and neuronal subtypes regulating chronic pain. Nature neuroscience, 26(11), 1880-1893 (2023).
22. He, Y., Lu, W., Zhou, X., Mu, J., et al. Unraveling Alzheimer’s disease: insights from single-cell sequencing and spatial transcriptomic. Frontiers in Neurology, 1515981 (2024).
23. Shimogori, T., Onishi, K., Hoshino, T., et al. Molecular architecture of primate specific neural circuit formation. Research Square, rs-3. (2024).
24. Gao, S., Shi, Q., Zhang, Y., et al. Identification of HSC/MPP expansion units in fetal liver by single-cell spatiotemporal transcriptomics. Cell research, 32, 38-53 (2022).
25. Holler, K., Neuschulz, A., Drewe-Boß, P., et al., Spatio-temporal mRNA tracking in the early zebrafish embryo. Nature communications, 12, 3358 (2021).
26. Maciejewski, K., & Czerwinska, P. Scoping Review: Methods and Applications of Spatial Transcriptomics in Tumor Research. Cancers, 16, 3100 (2024).
27. Toyama, S., Honda, T., Iwabuchi, S., et al. Application of spatial transcriptomics analysis using the Visium system for the mouse nasal cavity after intranasal vaccination. Frontiers in Immunology, 14, 1209945 (2023).
28. Miyamoto, A. T., Shimagami, H., Kumanogoh, A., et al., Spatial transcriptomics in autoimmune rheumatic disease: potential clinical applications and perspectives. Inflammation and Regeneration, 45, 6 (2025).
29. Zhang, L., Chen, D., Song, D., et al., Clinical and translational values of spatial transcriptomics. Signal Transduction and Targeted Therapy, 7, 111 (2022).
30. Fortner, A., & Bucur, O. Multiplexed spatial transcriptomics methods and the application of expansion microscopy. Frontiers in Cell and Developmental Biology, 12, 1378875 (2024).
31. Lee, J., Yoo, M., & Choi, J. Recent advances in spatially resolved transcriptomics: challenges and opportunities. BMB reports, 55, 113 (2022).
32. Si, Z., Li, H., Shang, W., Zhao, Y., et al. SpaNCMG: improving spatial domains identification of spatial transcriptomics using neighborhood-complementary mixed-view graph convolutional network. Briefings in Bioinformatics, 25, 259 (2024).
33. Chen, C., Ge, Y., & Lu, L. Opportunities and challenges in the application of single-cell and spatial transcriptomics in plants. Frontiers in Plant Science, 14, 1185377 (2023).
34. Lebrigand K., Bergenstråhle J., Thrane K, et al. The spatial landscape of gene expression isoforms in tissue sections. Nucleic Acids Res, 51, (2023).