Spatial Transcriptomics Workflow: From Sample Preparation to Data Analysis

Spatial transcriptomics preserves the positional context of gene expression within intact tissue, allowing researchers to map where transcripts originate instead of averaging them across a dissociated cell suspension. This workflow guide covers the complete experimental pipeline from tissue sectioning and permeabilization through library construction, sequencing, and computational analysis, with quality control checkpoints at each stage. The intended reader is a student, researcher, analyst, or life-science professional who needs a practical operational reference for designing and executing spatial transcriptomics experiments.

What Spatial Transcriptomics Measures and Why Position Matters

Cells are the fundamental units of biological systems and exhibit unique development trajectories and molecular features. Single-cell RNA sequencing technologies have expanded horizontally to include genome, epigenome, proteome, and metabolome measurements, while vertical integration now incorporates spatial transcriptomics and CRISPR screening [5]. The central limitation of conventional single-cell sequencing is the loss of spatial information during cell preparation, which prevents researchers from knowing where each cell resided within the tissue architecture [20]. Spatial transcriptomics addresses this gap by capturing gene expression data while retaining the two-dimensional or three-dimensional coordinates of each measurement point.

The biological rationale for preserving spatial context is straightforward. Tissue function depends on cellular neighborhoods, gradients, and cell-to-cell signaling that cannot be reconstructed from a dissociated cell suspension. Imaging-based spatial transcriptomics techniques characterize gene expression in cells within their native context by imaging barcoded probes for mRNA at single-molecule resolution [8]. This positional information allows researchers to ask questions about cellular microenvironments, tissue organization, and disease pathology that are inaccessible to bulk or single-cell approaches.

Spatial transcriptomics has wide-ranging applications in biological research and bioinformatics, allowing researchers to investigate and track spatial variations in gene expression across different tissues, conditions, and diseases [12]. In cancer research, spatial transcriptomics preserves the spatial information of RNA transcripts, facilitating a deeper understanding of tumor heterogeneity and the intricate interplay between tumor cells and the tumor microenvironment [20]. In plant biology, spatial approaches combined with droplet-based single-cell technologies have enhanced understanding of complex biological processes, though rigid cell walls and size variability require adaptation of mammalian-derived analytical methods [23].

Core Principles of Spatial Transcriptomics Experimental Design

Platform Selection Determines Resolution and Throughput

Spatial transcriptomics platforms fall into two broad categories: sequencing-based and imaging-based. Sequencing-based platforms capture mRNA from tissue sections onto spatially barcoded arrays, where each capture spot carries a unique positional barcode. Imaging-based platforms use fluorescence in situ hybridization or in situ sequencing to visualize individual transcripts directly within the tissue. The choice between these approaches depends on the biological question, sample type, budget, and available infrastructure.

Imaging-based spatial transcriptomics has advanced from low-plex single-molecule fluorescence in situ hybridization to a diverse set of highly multiplexed platforms, with recent multimodal and pathology-compatible capabilities [18]. These platforms differ in chemistry, coding, and imaging strategies, yet their biological interpretation converges on common computational problems including preprocessing, registration, restoration, feature detection, barcode decoding, molecule calling, cell segmentation, transcript assignment, probabilistic cell typing, spatial-domain inference, and atlas integration [18].

Sequencing-based platforms such as Visium capture polyadenylated mRNA on spatially barcoded slides. The practical guide informed by processing and analysis of over 1000 spatial samples across multiple ST platforms emphasizes that platform selection, sample quality, and experimental scalability are the primary practical barriers to implementation [9]. Researchers designing their first spatial experiment should consider whether spot-level resolution is sufficient or whether single-cell resolution is required, as this decision drives platform choice.

Sample Quality Determines Data Quality

The single most important determinant of spatial transcriptomics data quality is the quality of the input tissue. RNA integrity, tissue preservation, and section thickness all directly affect the number of transcripts captured and the reliability of downstream analysis. The practical guide from over 1000 samples identifies sample quality as one of the primary practical barriers to implementation [9].

Fresh frozen tissue is the standard input for most spatial transcriptomics platforms. Tissue should be embedded in optimal cutting temperature compound and stored at minus 80 degrees Celsius immediately after collection. The time between tissue harvest and freezing should be minimized to prevent RNA degradation. Clinical samples present additional challenges because surgical specimens may have prolonged warm ischemia times before freezing, which degrades RNA quality [9].

For imaging-based approaches, tissue preparation must also preserve morphology for accurate cell segmentation. Single-molecule fluorescence in situ hybridization requires sample preparation that overcomes hybridization and imaging challenges, including fixation, permeabilization, and probe hybridization steps [6]. The workflow for imaging-based spatial transcriptomics on retinal flatmounts required optimization of the combined imaging-based spatial transcriptomics and immunostaining workflow, demonstrating that tissue-specific protocol optimization is often necessary [7].

Experimental Design Requires Statistical Power Considerations

Gene expression profiling technologies produce data with distinct characteristics, and guaranteeing biologically meaningful findings requires systematic consideration of experimental factors through statistical power analysis [19]. For bulk RNA-seq and single-cell RNA-seq, power analysis tools exist for each research objective. For high-throughput spatial transcriptomics, no dedicated power analysis tools are currently available, so researchers must instead investigate the factors that can influence power [19].

Key factors influencing statistical power in spatial transcriptomics include the number of biological replicates, the number of capture spots or cells per sample, sequencing depth, and the magnitude of the biological effect being measured. Spatial data have inherent spatial autocorrelation, meaning neighboring spots are more similar than distant spots, which affects the effective sample size for statistical tests. Researchers should consult the platform-specific recommendations for sequencing depth and account for the spatial structure of the data when planning replicate numbers.

At a Glance: Spatial Transcriptomics Workflow Decision Table

Practical Workflow: From Tissue to Data

Step 1: Tissue Collection and Quality Assessment

The first decision in any spatial transcriptomics experiment is the source and condition of the tissue. Fresh frozen tissue is the standard input for most platforms. Collect tissue immediately after harvest, embed in optimal cutting temperature compound, and freeze in isopentane cooled on dry ice or liquid nitrogen. Record the time from harvest to freezing for every sample because this metric correlates with RNA quality.

Before sectioning, assess RNA quality from a representative tissue piece. Extract RNA and measure the RNA integrity number using a bioanalyzer or similar instrument. Samples with low RNA integrity will produce poor spatial transcriptomics data regardless of downstream optimization. The practical guide from over 1000 samples emphasizes that sample quality is a primary practical barrier, with special attention needed for clinical samples [9].

For plant tissues, additional considerations apply. Plant cells have rigid cell walls and variable sizes that require adaptation of mammalian-derived analytical methods [23]. Plant single-cell and spatial transcriptomics require careful consideration of tissue fixation, cell wall digestion, and protoplasting steps [11]. Researchers working with plant tissues should consult plant-specific protocols instead of assuming mammalian protocols transfer directly.

Step 2: Sectioning and Mounting

Cryosection the embedded tissue at the recommended thickness for the chosen platform. Typical section thickness ranges from 5 to 20 micrometers depending on the platform and tissue type. Sections must be flat, free of folds, and firmly adhered to the capture surface.

Tissue adherence is a common failure point. Sections that detach during permeabilization or library preparation lose spatial information and contaminate adjacent capture areas. Optimize sectioning temperature and use charged or adhesive slides as recommended by the platform manufacturer. Inspect every section microscopically before proceeding to permeabilization.

For imaging-based platforms, section thickness affects optical clarity and the ability to resolve individual transcripts. Thicker sections increase the amount of out-of-focus fluorescence and complicate cell segmentation. The computational difficulty of imaging-based spatial transcriptomics increases with optical crowding and tissue thickness [18]. Use the thinnest section that preserves tissue morphology and contains sufficient RNA for detection.

Step 3: Fixation and Permeabilization

Fixation preserves tissue morphology and prevents RNA degradation during the experiment. Most spatial transcriptomics workflows use methanol fixation or formalin fixation followed by permeabilization. The permeabilization step creates pores in the cell membranes that allow mRNA to diffuse to the capture surface or hybridize to probes.

Permeabilization time is the most critical optimization parameter for sequencing-based platforms. Too little permeabilization results in low RNA capture, while too much permeabilization causes RNA to diffuse away from its origin, blurring the spatial signal. Optimize permeabilization time using test sections and a fluorescent reverse transcription reaction to visualize RNA capture efficiency.

The practical guide from over 1000 samples emphasizes that experimental scalability depends on consistent tissue handling and permeabilization across batches [9]. When processing multiple samples, use the same permeabilization time for all samples of the same tissue type. Different tissue types may require different permeabilization times because of differences in extracellular matrix density and cell membrane composition.

Step 4: Library Construction

Library construction converts captured mRNA into a sequencing-ready library. For sequencing-based platforms, this involves reverse transcription on the capture surface, second strand synthesis, and amplification. The spatial barcode is incorporated during reverse transcription, linking each transcript to its positional coordinate.

Amplification cycles must be optimized to produce sufficient library yield without introducing amplification bias. Overamplification can create duplicate reads that consume sequencing capacity without adding biological information. Quantify the library using quantitative PCR and assess fragment size distribution using a bioanalyzer or similar instrument.

For imaging-based platforms, library construction is replaced by probe hybridization and signal amplification. Multiplexed error-robust fluorescence in situ hybridization uses barcoded probes that are read across multiple imaging rounds [8]. The number of imaging rounds determines the number of genes that can be detected, with more rounds enabling higher multiplexing at the cost of increased imaging time and data volume.

Step 5: Sequencing

Sequencing-based spatial transcriptomics libraries are sequenced on high-throughput platforms. The required read depth depends on the platform, the number of capture spots, and the biological question. Deeper sequencing increases the sensitivity for detecting lowly expressed genes but also increases cost.

The Joint Sparse method for Imaging Transcriptomics demonstrates that lower magnification imaging data can be decoded using optimization algorithms that incorporate codebook knowledge and sparsity assumptions, reducing the need for high-magnification imaging [8]. This approach improves throughput and recovery performance over standard decoding methods, suggesting that imaging parameters can be adjusted to balance resolution and throughput [8].

For sequencing-based platforms, monitor sequencing quality metrics including per-base quality scores, mapping rates, and duplication rates. Low mapping rates may indicate contamination or poor library quality. High duplication rates suggest overamplification during library construction.

Step 6: Data Analysis

Spatial transcriptomics data analysis involves multiple stages from raw data processing to biological interpretation. The analytical framework converts raw fluorescence signals or sequencing reads into molecule-level, cell-level, and tissue-level representations [18]. The specific steps depend on the platform and the biological question.

The STAT multi-agent framework demonstrates that spatial transcriptomics analysis often involves a myriad of computational methods across diverse platforms, leading analysts to spend excessive time on data assembly instead of deriving biological insights [17]. Integrated analysis platforms can reduce this burden by providing staged pipelines with interactive visualization [17].

The spatialGE software provides visualizations and quantification of tumor microenvironment heterogeneity through gene expression surfaces, spatial heterogeneity statistics, spot-level cell deconvolution, and spatially informed clustering [22]. Tools like spatialGE enable researchers to explore associations between spatial heterogeneity and clinical data [22].

Computational Analysis Pipeline

Preprocessing and Quality Control

The first computational step is preprocessing raw data into a count matrix where each row represents a gene and each column represents a capture spot or cell. For sequencing-based platforms, this involves demultiplexing, read alignment, and assignment of reads to spatial barcodes. For imaging-based platforms, preprocessing involves image registration, spot detection, and barcode decoding.

Quality control metrics for spatial transcriptomics data include the number of genes detected per spot, the number of transcripts per spot, and the fraction of mitochondrial reads. Spots with very low transcript counts may represent tissue folds or areas of poor permeabilization. Spots with very high mitochondrial fractions may represent dying cells or areas of tissue damage.

The practical guide from over 1000 samples outlines best practices for computational analysis, emphasizing the importance of quality control visualization before downstream analysis [9]. Visualize the spatial distribution of quality metrics to identify problematic regions of the tissue section.

Normalization and Batch Effect Correction

Normalization adjusts for differences in sequencing depth and capture efficiency across spots. Common approaches include library size normalization, where each spot is scaled to a common total transcript count, and variance stabilizing transformations. The choice of normalization method affects downstream analyses including clustering and differential expression.

Batch effects arise when samples are processed in different batches or on different slides. These technical artifacts can obscure biological signal and create false associations. Batch effect correction methods developed for single-cell RNA sequencing can be adapted for spatial transcriptomics, but spatial information should be preserved during correction.

The Annotated Research Context framework demonstrates how structured metadata and ontology annotations ensure machine readability and interpretability of spatial transcriptomics data [15]. Capturing experimental metadata in standardized formats enables comparative analyses across different datasets and improves reproducibility [15].

Dimensionality Reduction and Clustering

Spatial transcriptomics data are high dimensional, with thousands of genes measured across thousands of spots. Dimensionality reduction methods including principal component analysis and uniform manifold approximation and projection project the data into a lower dimensional space that captures the major sources of variation.

Clustering groups spots or cells with similar gene expression profiles. The Leiden algorithm is commonly used for clustering single-cell and spatial data [21]. Spatially informed clustering methods incorporate spatial coordinates into the clustering objective, producing clusters that are contiguous in space [22].

The spatialGE STclust method performs spatially informed clustering of spots, producing clusters that reflect both transcriptional similarity and spatial proximity [22]. This approach is particularly useful for identifying tissue domains and anatomical structures.

Cell Type Decomposition and Annotation

Most sequencing-based spatial transcriptomics platforms measure gene expression at spot resolution, where each spot may contain multiple cells. Cell type decomposition estimates the proportion of each cell type within each spot using reference signatures from single-cell RNA sequencing data.

Cell type annotation assigns biological identities to clusters based on marker gene expression. The OmiCLIP visual-omics foundation model links hematoxylin and eosin images and transcriptomics using tissue patches from Visium data, enabling tissue alignment, annotation, cell-type decomposition, and spatial gene expression prediction from histology images [25].

For imaging-based platforms with single-cell resolution, cell segmentation assigns transcripts to individual cells based on nuclear and cytoplasmic staining. Probabilistic cell typing methods assign each cell to a cell type based on its transcript profile [18].

Spatial Domain Identification and Differential Expression

Spatial domains are regions of the tissue with distinct gene expression programs. Identifying spatial domains reveals tissue architecture and disease-associated niches. Methods for spatial domain identification include spatially informed clustering, hidden Markov random field models, and graph neural networks.

Differential expression analysis identifies genes that are upregulated or downregulated between spatial domains or between conditions. Spatial differential expression methods account for spatial autocorrelation, where neighboring spots are more similar than distant spots.

Geographically weighted regression, a spatial statistical method, has been applied to Visium spatial transcriptomics data to characterize spatially resolved high-coupling spots where M2 macrophage and fibrosis coupling is significantly positive in diabetic kidney disease [13]. This approach identified high-coupling spots enriched for B cell and tertiary lymphoid structure-like immune signatures, suggesting that the regions captured biologically meaningful immune microenvironments [13].

Trajectory and Cell-Cell Communication Analysis

Trajectory analysis orders cells or spots along a developmental or disease progression continuum. Spatial trajectories incorporate spatial information to identify the direction of progression across the tissue. Machine learning and trajectory analysis have been used to segment and rank airspaces on a gradient of remodeling severity in pulmonary fibrosis, identifying compositional and molecular changes associated with progressive distal lung pathology [24].

Cell-cell communication analysis infers ligand-receptor interactions between neighboring cells or spots. Spatial communication analysis uses the physical proximity of cells to identify signaling interactions that occur in specific tissue niches. These analyses can reveal how cellular microenvironments impact cell function and disease progression.

Records and Measurements for Reproducible Experiments

Documentation Requirements

Reproducible spatial transcriptomics experiments require detailed documentation of every step from tissue collection to data analysis. Record the following information for each sample:

Tissue source, harvest time, and storage conditions. The time from harvest to freezing directly affects RNA quality and should be recorded for every sample. RNA integrity number measurements should be documented for each tissue block.

Sectioning parameters including section thickness, cryostat temperature, and section orientation. Tissue adherence issues should be noted because they affect data quality.

Permeabilization time and conditions. This parameter is optimized for each tissue type and should be recorded to enable comparison across batches.

Library construction parameters including amplification cycles and indexing strategy. Index contamination can create false signals and should be monitored.

Sequencing parameters including platform, read length, and sequencing depth. These parameters affect sensitivity and should be consistent across batches.

Quality Control Records

Quality control records should include both experimental and computational metrics. Experimental metrics include RNA integrity numbers, tissue morphology assessments, and permeabilization test results. Computational metrics include sequencing quality scores, mapping rates, and per-spot quality distributions.

The FAIR Guiding Principles provide a framework for data management that ensures findability, accessibility, interoperability, and reusability [4]. Applying FAIR principles to spatial transcriptomics data involves depositing raw data in public repositories, providing structured metadata, and using standard file formats.

The Genomic Data Sharing Policy from the National Institutes of Health establishes expectations for sharing genomic data generated from NIH-funded research [3]. Researchers should review this policy when planning data sharing and deposition strategies.

Data Repositories and Sharing

Public repositories for spatial transcriptomics data include the NCBI Data Resources, which provides access to gene expression data through the Gene Expression Omnibus [2]. The EMBL-EBI Training program offers resources for learning about data submission and analysis [1].

The DeepSpaceDB database provides a comprehensive and dynamic resource for spatial transcriptomics data exploration, enabling analysis of quality indicators, spatially variable genes and pathways, and gene expression variations between anatomical regions [12]. Databases like DeepSpaceDB facilitate access to spatial data for researchers who cannot generate their own datasets due to technical and financial constraints [12].

Common Failure Patterns and Troubleshooting

Low RNA Quality

Low RNA quality is the most common cause of failed spatial transcriptomics experiments. Symptoms include low transcript counts per spot, poor spatial signal, and high background noise. The practical guide from over 1000 samples identifies sample quality as a primary practical barrier, with special attention to clinical samples [9].

Prevention strategies include minimizing warm ischemia time, rapid freezing, and RNA integrity assessment before sectioning. If RNA quality is marginal, consider whether the biological question can be answered with lower sensitivity or whether sample collection can be improved.

Poor Tissue Adherence

Tissue sections that detach from the capture surface during permeabilization or library preparation lose spatial information and contaminate adjacent areas. Symptoms include missing regions in the spatial data and high background signal.

Prevention strategies include optimizing sectioning temperature, using charged slides, and minimizing handling of the slides during the experiment. If adherence is consistently poor, test different slide types or coating conditions.

Suboptimal Permeabilization

Incorrect permeabilization time produces either low RNA capture or spatial diffusion of transcripts. Low capture appears as uniformly low transcript counts. Spatial diffusion appears as blurring of spatial patterns and loss of sharp boundaries between tissue regions.

Optimization involves testing a range of permeabilization times on test sections and visualizing the results. The optimal time balances capture efficiency against spatial resolution.

Batch Effects

Samples processed in different batches show systematic differences that can obscure biological signal. Symptoms include clustering by batch instead of by biological condition and differential expression results that reflect technical artifacts.

Prevention strategies include processing all samples in a single batch when possible, using consistent protocols across batches, and including technical replicates. Batch effect correction methods can mitigate but not eliminate batch effects.

Computational Pipeline Errors

Errors in the computational pipeline can produce misleading results. Common errors include incorrect reference genome selection, misassignment of spatial barcodes, and inappropriate normalization. The STAT multi-agent framework addresses these challenges by providing a staged skill-aware pipeline with persistent session and interactive tissue viewer [17].

Validation strategies include examining quality control plots, comparing results across analysis methods, and validating key findings with orthogonal approaches such as immunohistochemistry or in situ hybridization.

Limitations and Interpretation Boundaries

Resolution Limits

Sequencing-based spatial transcriptomics platforms measure gene expression at spot resolution, where each spot may contain multiple cells. This limits the ability to resolve cell-type-specific expression patterns within mixed spots. Cell type decomposition methods estimate cell type proportions but cannot fully recover single-cell expression profiles.

Imaging-based platforms achieve single-cell or subcellular resolution but are limited in the number of genes that can be detected. Panel design determines which genes are measured, and genes not included in the panel cannot be analyzed.

Sensitivity Limits

Spatial transcriptomics has lower sensitivity than bulk RNA sequencing for detecting lowly expressed genes. The capture efficiency of sequencing-based platforms is typically lower than single-cell RNA sequencing, and imaging-based platforms may miss transcripts below the detection threshold.

The Joint Sparse method for Imaging Transcriptomics demonstrates that decoding algorithms can improve recovery performance from lower magnification imaging data [8]. However, sensitivity limits remain a fundamental constraint of spatial technologies.

Technical Artifacts

Spatial transcriptomics data contain technical artifacts that can be mistaken for biological signal. These include batch effects, spatial gradients in capture efficiency, and tissue-specific artifacts such as autofluorescence in imaging-based platforms.

The review of imaging-based spatial transcriptomics highlights how optical crowding, tissue thickness, panel bias, and multimodal complexity increase computational difficulty [18]. Researchers should be aware of these artifacts when interpreting results.

Interpretation Boundaries

Spatial transcriptomics measures RNA abundance, not protein expression or protein localization. RNA abundance does not always correlate with protein abundance due to post-transcriptional regulation. Validation with protein-based methods is recommended for key findings.

Spatial transcriptomics provides a snapshot of gene expression at a single time point. Dynamic processes such as cell migration or signaling cannot be directly observed. Lineage tracing or time series experiments are needed to study dynamic processes.

Safety and Regulatory Context

Biosafety Considerations

Spatial transcriptomics experiments involve handling human and animal tissues that may contain infectious agents. Follow institutional biosafety guidelines for tissue handling, fixation, and disposal. Formalin fixation inactivates most infectious agents but does not eliminate all risks.

Chemical safety considerations include handling of optimal cutting temperature compound, fixatives, and organic solvents used in library preparation. Follow material safety data sheet recommendations for personal protective equipment and ventilation.

Data Privacy and Genomic Data Sharing

Spatial transcriptomics data from human samples contain genomic information that may be identifiable. The Genomic Data Sharing Policy from the National Institutes of Health establishes expectations for data sharing and privacy protection [3]. Researchers should review this policy and institutional review board requirements when planning studies with human samples.

De-identification of spatial transcriptomics data is challenging because spatial information may reveal individual-specific features. Consult institutional privacy officers when planning data sharing for human samples.

Ethical Considerations

Spatial transcriptomics studies using human tissue require appropriate ethical approval and informed consent. Researchers should ensure that tissue collection complies with institutional and national regulations.

The NCBI Data Resources provides access to genomic data with varying levels of access control [2]. Researchers should understand the data use restrictions for any dataset they access or deposit.

Professional Escalation Criteria

When to Seek Expert Consultation

Several situations warrant consultation with spatial transcriptomics experts or core facility staff:

If RNA integrity numbers are consistently low despite optimized collection procedures, consult with the pathology department or biorepository about collection protocols.

If permeabilization optimization fails to produce adequate capture, consult the platform manufacturer or an experienced user for troubleshooting.

If computational results are inconsistent across analysis methods, consult a bioinformatics core facility for guidance on pipeline selection and parameter optimization.

If clinical samples are involved, consult with the clinical team about sample collection logistics and ethical requirements.

When to Abort an Experiment

Some situations indicate that an experiment should be aborted and restarted instead of continued:

If tissue sections show extensive folding or detachment during the experiment, the spatial information is compromised and results will be unreliable.

If library yield is far below the recommended range, the experiment may not produce sufficient sequencing data for meaningful analysis.

If sequencing quality metrics are poor, the data may not be interpretable and re-sequencing may be required.

When to Seek Medical or Veterinary Consultation

For studies involving animal models, consult with veterinary staff about tissue collection procedures and animal welfare requirements. For studies involving human clinical samples, consult with the clinical team about sample quality and clinical context.

The study of retinal ganglion cells on retinal flatmounts demonstrates the importance of tissue-specific protocol optimization for imaging-based spatial transcriptomics [7]. Researchers working with specialized tissues should seek guidance from experts with experience in those tissues.

Frequently Asked Questions

What is the difference between sequencing-based and imaging-based spatial transcriptomics?

Sequencing-based platforms capture mRNA onto spatially barcoded arrays, where each capture spot carries a unique positional barcode that is read during sequencing. Imaging-based platforms use fluorescence in situ hybridization or in situ sequencing to visualize individual transcripts directly within the tissue. Sequencing-based platforms typically measure the whole transcriptome at spot resolution, while imaging-based platforms measure a targeted gene panel at single-cell or subcellular resolution. The choice between approaches depends on whether whole-transcriptome coverage or single-cell resolution is more important for the biological question.

How much RNA quality is needed for spatial transcriptomics?

RNA quality directly affects the number of transcripts captured and the reliability of downstream analysis. The practical guide from over 1000 samples identifies sample quality as a primary practical barrier to implementation [9]. RNA integrity number assessment before sectioning is recommended, and samples with low RNA integrity will produce poor spatial transcriptomics data regardless of downstream optimization. The specific threshold depends on the platform and tissue type, so consult platform-specific recommendations.

How do I choose the right spatial transcriptomics platform for my experiment?

Platform selection depends on the biological question, sample type, budget, and available infrastructure. Consider whether spot-level resolution is sufficient or whether single-cell resolution is required. Consider whether whole-transcriptome coverage or a targeted gene panel is needed. Consider the number of samples and the scalability of the platform. The practical guide from over 1000 samples emphasizes that platform selection, sample quality, and experimental scalability are the primary practical barriers to implementation [9].

What is the optimal permeabilization time for my tissue?

Permeabilization time must be optimized for each tissue type because differences in extracellular matrix density and cell membrane composition affect RNA diffusion. Test a range of permeabilization times on test sections and visualize the results using a fluorescent reverse transcription reaction. The optimal time balances capture efficiency against spatial resolution. Too little permeabilization results in low RNA capture, while too much permeabilization causes RNA to diffuse away from its origin.

How many biological replicates do I need for a spatial transcriptomics experiment?

The number of biological replicates depends on the biological variability of the system, the magnitude of the effect being measured, and the statistical power required. No dedicated power analysis tools exist for high-throughput spatial transcriptomics, so researchers must investigate the factors that can influence power [19]. Spatial data have inherent spatial autocorrelation, which affects the effective sample size for statistical tests. Consult a biostatistician when planning replicate numbers.

What quality control metrics should I examine for spatial transcriptomics data?

Examine the number of genes detected per spot, the number of transcripts per spot, and the fraction of mitochondrial reads. Visualize the spatial distribution of quality metrics to identify problematic regions of the tissue section. For sequencing-based platforms, monitor sequencing quality metrics including per-base quality scores, mapping rates, and duplication rates. The practical guide from over 1000 samples outlines best practices for quality control visualization before downstream analysis [9].

How do I validate spatial transcriptomics findings?

Validate key findings with orthogonal approaches such as immunohistochemistry, in situ hybridization, or quantitative PCR. Single-molecule fluorescence in situ hybridization is a powerful method for the visualization and quantification of individual RNA molecules within intact cells, offering validation at single-cell and single-molecule resolution [6]. Validation is particularly important for findings that drive biological conclusions or clinical implications.

What are the main limitations of spatial transcriptomics?

Spatial transcriptomics has lower sensitivity than bulk RNA sequencing for detecting lowly expressed genes. Sequencing-based platforms measure gene expression at spot resolution, where each spot may contain multiple cells. Imaging-based platforms are limited in the number of genes that can be detected. Spatial transcriptomics measures RNA abundance, not protein expression. Technical artifacts including batch effects and spatial gradients in capture efficiency can be mistaken for biological signal.

Related Bioinformatics Guides

• Alternative Splicing Analysis from RNA-Seq Data
• Spatial Transcriptomics Alignment and Cellular Neighborhood Analysis
• RNA Seq Fastq Example: Structural Analysis and Computational Methodologies in Bioinformatics
• Modern Transcriptomics: From Bulk RNA-Seq to Single-Cell and Spatial Resolution
• Circular RNAs: Computational Identification and Analysis

References and Further Reading

• EMBL-EBI Training. European Bioinformatics Institute.
• NCBI Data Resources. National Center for Biotechnology Information.
• Genomic Data Sharing Policy. National Institutes of Health.
• The FAIR Guiding Principles. Scientific Data.
• Single-cell omics: experimental workflow, data analyses and applications.. Science China. Life sciences, 2025.
• smFISH for Plants.. Methods in molecular biology (Clifton, N.J.), 2024.
• Molecular and spatial analysis of ganglion cells on retinal flatmounts identifies perivascular neurons resilient to glaucoma.. Neuron, 2025.
• Optimization-based decoding of Imaging Spatial Transcriptomics data.. Bioinformatics (Oxford, England), 2023.
• A practical guide to spatial transcriptomics: lessons from over 1000 samples.. Trends in biotechnology, 2026.
• Next-generation spatial transcriptomics: unleashing the power to gear up translational oncology.. MedComm, 2024.
• Plant Single-Cell/Nucleus RNA-seq Workflow.. Methods in molecular biology (Clifton, N.J.), 2023.
• Mining Spatial Transcriptomics Datasets using DeepSpaceDB.. Journal of visualized experiments : JoVE, 2025.
• Applying Spatial Statistics to Spatial Transcriptomics Reveals Local Association Between M2-like Macrophages and Fibrosis in Diabetic Kidney Disease. 2026.
• Editorial: AI in single-cell biology
• Integrating cross-omics research through FAIR Digital Objects with DataPLANT.. 2026.
• Progress in the Application of Machine Learning in the Field of Single-Cell and Spatial Transcriptomics. 2026.
• STAT: A multi-agent framework for integrated and interactive spatial transcriptomics analysis. 2026.
• Imaging-Based Spatial Transcriptomics: Data Interpretation Methods and Biomedical Applications.. 2026.
• Statistical Power Analysis for Designing Bulk, Single-Cell, and Spatial Transcriptomics Experiments: Review, Tutorial, and Perspectives. Biomolecules, 2023.
• Spatial transcriptomics: a new frontier in cancer research. Clinical Cancer Bulletin, 2024.
• Incorporating Practical Single Cell and Spatial Transcriptomics Analysis in a Bioinformatics Course. 2024 11th International Conference on Electrical, Electronic and Computing Engineering (IcETRAN), 2024.
• spatialGE: Quantification and visualization of the tumor microenvironment heterogeneity using spatial transcriptomics. bioRxiv, 2021.
• Advances in plant spatial multi-omics data analysis.. Trends in Plant Science, 2025.
• Spatial transcriptomics identifies molecular niche dysregulation associated with distal lung remodeling in pulmonary fibrosis. Nature Genetics, 2025.
• A visual-omics foundation model to bridge histopathology with spatial transcriptomics. Nature Methods, 2025.
• Mapping biology in space: from spatial transcriptomics platforms to analytical tools and databases. Science Bulletin, 2026.
• CCMAS: A Multi-Agent System for Identification and Functional Interpretation of Cellular Community Motifs in Spatial Transcriptomics. 2026 5th International Conference on Electronics Technology and Artificial Intelligence Etai 2026, 2026.
• Spatial transcriptomics reveals strong association between SFRP4 and extracellular matrix remodeling in prostate cancer. Communications Biology, 2024.

This article is educational and does not replace validated analysis plans, institutional policy, clinical interpretation, or specialist review.