Sample Prep to Data Analysis: A Workflow Guide to Spatial Transcriptomics

Spatial transcriptomics is changing how we study tissue biology. By preserving the spatial context of gene expression, this method reveals not just which genes are active, but exactly where they’re expressed within intact tissue. From cancer microenvironments to brain development, the technology offers a powerful lens into cellular function and organization.

That said, it’s a complex workflow. Every step—from tissue handling to data processing—can impact results. This guide outlines the typical end-to-end process researchers follow to generate reliable, interpretable spatial transcriptomics data.

1.  Tissue Preparation

High-quality results begin with proper tissue handling. Most platforms require either fresh-frozen or FFPE-preserved samples. Tissues are sectioned at ~5–10 µm thickness and mounted on slides pre-coated with capture probes.

Key considerations:

• Maintain cold chain integrity to preserve RNA quality
• Avoid tissue folds or tearing during sectioning
• Store slides with care to prevent degradation

2.  Staining and Imaging

Before capturing RNA, sections are often stained with H&E or immunofluorescence. These images help researchers map gene expression data back to tissue architecture.

Ensure:

• Consistent imaging resolution across samples
• Clean slides without background artifacts
• Imaging occurs before any enzymatic steps

3.  RNA Capture and Library Construction

Once slides are imaged, permeabilization allows mRNA to bind to barcoded probes on the slide surface. These location-specific barcodes are preserved during reverse transcription and subsequent cDNA amplification.

Critical to success:

• Optimize permeabilization time for tissue type
• Perform quality checks on cDNA yield and integrity
• Use clean reagents to avoid background signal

4.  Sequencing

Libraries are sequenced using high-throughput short-read platforms. Coverage requirements vary based on the desired resolution and biological complexity but often range between 50K–100K reads per spot.

Recommendations:

• Follow vendor guidance on read depth and layout
• Pair-end sequencing helps with alignment accuracy
• Include sequencing controls if comparing across batches

5.  Data Processing

Following sequencing, raw reads are demultiplexed, aligned to the reference genome, and linked back to their spatial barcodes. Software like 10x Genomics’ Space Ranger or NanoString’s GeoMx DSP tools automate much of this step.

Outputs typically include:

• A spatially resolved expression matrix
• Quality metrics for each capture spot
• Integrated histology overlays

6.  Analysis and Visualization

With the spatial expression matrix in hand, researchers can begin interpreting the data:

• Cluster regions based on expression patterns
• Identify differentially expressed genes between tissue zones
• Overlay gene expression with cell morphology
• Use reference scRNA-seq data for cell type deconvolution

In more advanced cases, spatial data can be integrated with proteomics, machine learning models, or multi-tissue comparisons.

7.  Use Cases and Access

Spatial transcriptomics has broad applications in cancer research, neuroscience, infectious disease, and developmental biology. It’s now supported by several commercial platforms and service providers. For an overview of how this technology works in practice, including common platforms and sequencing protocols, explore this resource on spatial transcriptomics.

Final Thoughts

As spatial methods become more accessible, it’s increasingly important to understand the full workflow and its nuances. Thoughtful preparation, platform familiarity, and careful data handling can significantly impact the success of a spatial transcriptomics project. Whether in-house or through a research partner, mastering this workflow opens the door to deeper, more contextual insights into complex biology.

Let me know if you’d like a visual workflow diagram or version adapted for email or LinkedIn distribution.