View on GitHub

Transcriptomics

Fundamentals of Transcriptomics — from RNA sequencing basics to advanced expression analysis.

Spatial Transcriptomics

“In biology, where a gene is expressed is as important as what is expressed.”

1. Introduction to Spatial Transcriptomics

Spatial transcriptomics (ST) is a technique that maps gene expression profiles within the intact context of a tissue, preserving the spatial organization of cells and their molecular states. Unlike traditional transcriptomics methods, ST combines high-throughput RNA sequencing or imaging with spatial coordinates, allowing researchers to visualize how genes are expressed in specific locations and how cells interact within their microenvironment. This is achieved by slicing tissue samples, treating them to release RNA that binds to barcoded spots on a slide, sequencing the barcoded RNA, and integrating this data with imaging to associate transcripts with precise positions. The core principle builds on single-cell genomics, using histological staining and microscopy to link RNA data to cellular and tissue architecture.

Importance in Biology

ST is transformative for understanding multicellular organisms, where spatial organization drives organ function, development, and disease. It reveals cellular heterogeneity, cell-cell interactions, and microenvironmental dynamics that are lost in dissociated samples. By providing context to gene expression—such as how immune cells infiltrate tumors or how injury propagates in tissues—ST bridges historical histology with modern omics, accelerating discoveries in developmental biology, immunology, and pathology. It is particularly vital for studying complex diseases like cancer and neurodegeneration, informing drug targeting, precision medicine, and tissue engineering.

Key Concept Description Biological Relevance
Spatial Coordinates GPS-like mapping of transcripts to tissue locations. Enables neighborhood analysis (e.g., tumor-immune interfaces).
Transcript Capture Barcoded probes or arrays bind RNA in situ. Captures whole transcriptome or targeted genes without dissociation.
Multi-Modal Integration Combines RNA with proteins, epigenomics. Provides holistic views of cellular states.

2. Technologies and Methods

ST technologies are categorized into imaging-based (using fluorescence in situ hybridization for targeted detection) and sequencing-based (using barcoded arrays for unbiased profiling). Imaging-based methods excel in high-resolution, targeted studies, while sequencing-based offer broader coverage but lower resolution.

Imaging-Based Technologies

These rely on single-molecule FISH (smFISH) for subcellular precision, suitable for hypothesis-driven research.

Technology Resolution Throughput Pros Cons
Xenium (10x Genomics) ~1 μm (subcellular) Low (hours-days) High sensitivity/specificity; FFPE-compatible; up to 6,000 genes; custom panels. Targeted only; long imaging; no protein co-detection; high custom cost.
MERSCOPE (Akoya Biosciences ~1 μm (subcellular) Low Customizable (up to 1,000 genes); RNA/protein co-detection; FFPE/cells. Gene limit; long times; lower FFPE sensitivity.
CosMx SMI (NanoString) ~1 μm (subcellular) Low Up to 6,000 genes; RNA/protein co-detection; fast cycles; FFPE. Lower specificity; sequential imaging; costly.

Sequencing-Based Technologies

These use next-generation sequencing for whole-transcriptome analysis, ideal for discovery.

Technology Resolution Throughput Pros Cons
Visium (10x Genomics) ~55 μm (multi-cell) High Whole transcriptome; FFPE/fresh; large area; protein panels. No single-cell; low sensitivity for rare transcripts; needs deconvolution.
Visium HD (10x Genomics) ~2 μm (near single-cell) Medium High resolution; whole transcriptome; human/mouse FFPE. Reduced genes at high res; species-limited; smaller area.
Stereo-seq ~0.2 μm (near single-cell) Medium Nanoscale res; whole transcriptome; all species/FFPE; large area. Diffusion artifacts; specialized sequencer; lower sensitivity.
GeoMx DSP (NanoString) ~50 μm (ROI-based) High Targeted ROI; RNA/protein; FFPE human/mouse; multi-sample. Predefined ROIs; no single-cell; species-limited.

Guidance for Selection

3. Comparison to Bulk and Single-Cell RNA-Seq

ST bridges the gap between bulk RNA-seq (averaged profiles) and single-cell RNA-seq (scRNA-seq; cellular resolution without space). Below is a comparison:

Feature Bulk RNA-Seq Single-Cell RNA-Seq (scRNA-seq) Spatial Transcriptomics (ST)
Methodology Pools RNA from entire sample; mRNA/whole transcriptome sequencing. Isolates cells (e.g., 10x Chromium droplets with barcodes/UMIs); sequences individual profiles. Barcoded arrays/slides capture RNA in situ; sequences with spatial mapping (e.g., Visium probes).
Resolution Population-level average. Single-cell. Spatial (subcellular to multi-cell; e.g., 1-55 μm spots).
Advantages Cost-effective; detects fusions/biomarkers; clinical (diagnosis/prognosis). Resolves heterogeneity/rare cells; microenvironment analysis. Preserves tissue architecture; cell interactions; multimodal (RNA/protein).
Disadvantages Masks heterogeneity; sampling bias. Loses spatial info; dissociation stress; low transcript recovery; costly. Suboptimal resolution; low sensitivity; high cost/labor.
Complementary Uses Foundational for broad profiling; pairs with scRNA-seq/ST for depth. Heterogeneity focus; integrates with ST for spatial restoration. Adds context to bulk/scRNA-seq; ideal for tumor microenvironments.

ST complements by restoring space lost in scRNA-seq, while offering more detail than bulk.

4. Applications in Biology and Disease

ST uncovers spatially resolved mechanisms, from development to pathology. #### Kidney Disease

Cancer

Neuroscience

Other Areas

ST excels in neighborhood mapping and ligand-receptor inference, aiding biomarker discovery.