2 Introduction
2.1 What is Spatial Trancriptomics?
Spatial transcriptomics refers to a group of technologies that enable the measurement of gene expression while preserving the spatial organization of tissue samples. This is a major advancement over traditional RNA sequencing, which loses spatial information during tissue dissociation.
By combining transcriptomic data with tissue morphology, spatial technologies help us answer questions such as:
- Where in the tissue are specific genes expressed?
- How do different cell types interact in their native environment?
- How does spatial context influence gene regulation in health and disease?
2.2 Platforms We’ll Cover
In this workshop, we will learn how to process data from three leading spatial transcriptomics platforms:
10x Genomics Visium
Whole-transcriptome capture on tissue sections with spatial barcoding and histology.NanoString CosMx
High-plex, single-molecule spatial profiling with subcellular resolution.10x Genomics Xenium
In situ RNA detection using padlock probes and imaging, offering near-single-cell resolution.
Each platform generates different types of data and requires tailored analysis approaches. By the end of this workshop, you will be familiar with:
- The structure of raw data from each platform
- Key steps in pre-processing and quality control
- How to extract and interpret spatially-resolved gene expression
2.3 Comparison of Spatial Transcriptomics Platforms
| Feature | Visium HD | Xenium | CosMx SMI |
|---|---|---|---|
| Platform Type | Capture-based sequencing | Imaging-based in situ detection | Imaging-based single-molecule FISH |
| Resolution | ~5–10 μm (subcellular bins) | Near single-cell / subcellular | Subcellular |
| Gene Panel | Whole transcriptome | Targeted (~400–500 genes) | Targeted (up to 1,000 genes) |
| Sample Type | FF or FFPE | FFPE | FFPE |
| Cell Segmentation | Approximate (binning) | Provided by pipeline | Based on nuclear/membrane stains |
| Spatial Barcode Approach | Barcoded slide (dense array) | Padlock probes + imaging | Hybridization + cyclic imaging |
| Readout | Sequencing | Fluorescence imaging | Fluorescence imaging |
| Throughput | High (whole tissue sections) | Medium (fields of view) | Medium (fields of view) |
| Strengths | Whole transcriptome + scale | High specificity, spatial detail | High plex, protein + RNA support |
| Analysis Software | Space Ranger, Seurat | Xenium Explorer, Seurat | CosMx SMI Pipeline, Squidpy |
Note: FF = Fresh Frozen, FFPE = Formalin-Fixed Paraffin-Embedded
2.4 What Research Questions Can Spatial Transcriptomics Answer?
Spatial transcriptomics enables researchers to explore biological questions that require both gene expression and spatial context. Some key research questions include:
2.4.1 Tissue Architecture & Cell Type Localization
- Where are different cell types located within a tissue?
- How do gene expression patterns vary across tissue regions (e.g., tumor core vs. margin)?
- Can we define spatial domains or regions based on transcriptomic profiles?
2.4.2 Cell-Cell Interactions & Microenvironments
- Which cells are interacting spatially and what genes are involved in these interactions?
- What ligands and receptors are co-expressed in adjacent cells?
- How do immune cells infiltrate tumors or inflamed tissues?
2.4.3 Disease Mechanisms & Pathology
- How does spatial gene expression change in disease vs. healthy tissue?
- What are the molecular signatures of diseased areas (e.g., fibrotic zones, plaques, tumors)?
- Can we detect early molecular changes before visible pathology?