Make Discoveries Happen.

Why Spatial Transcriptomics?

Spatial information is critical for understanding gene expression patterns within tissue and how individual cells interact with their surrounding environment.

In recent years, single-cell technology has gained favor over traditional bulk RNA approaches. Bulk RNA sequencing masks heterogeneity within a sample, which makes it difficult to characterize complex tissue such as tumors. Single-cell RNA sequencing allows for identification of rare or unique cell populations that are otherwise difficult or impossible to isolate.

However, today’s single-cell technologies require dissociation of tissue and subsequent loss of spatial information. Spatial transcriptomics allows you to acquire gene expression data at single-cell or near-single-cell resolution without losing spatial information.

Spatial transcriptomics methods are now being used to identify biomarkers and elucidate the mechanisms underlying disease. Spatial transcriptomics datasets can be integrated with other single-cell experiments to better characterize complex tissues.

Source: 10X Genomics

How it works

At Three Dimension Genomics, we use Visium, a whole-transcriptome spatial method developed by 10X Genomics. The Visium Spatial Gene Expression slide has 4 captures areas that each span 6.5mm x 6.5 mm and hold ~5,000 barcoded spots. Since an individual spot is 55um, you can expect to capture between 1-10 cells per spot depending on your tissue type.

Source: 10X Genomics

Tissue sections (either fresh-frozen or FFPE) are cut and placed onto the 10X Visium slide, which is then H&E stained (or stained for immunofluorescence) and imaged with a brightfield microscope. The spot’s spatial barcode is retained throughout library preparation and later used to visualize gene expression across the tissue.

Advantages of 10X Visium

10X Visium is compatible with both fresh-frozen and FFPE samples and has been optimized across a variety of human and animal tissues. When profiling fresh-frozen tissues, Visium is suitable for any species with a reference genome.

10X Visium has high resolution compared to other spatial transcriptomics technologies, allowing for identification of 1-10 cells per barcoded spatial area. It also has advantages over in situ hybridization-based approaches (such as single-molecule FISH), which require targeting to a limited number of mRNAs. 10X Visium, however, allows unbiased detection of mRNAs across the whole transcriptome.

Source: 10X Genomics

10X Genomics provides an interactive software called Loupe Browser that allows for easy navigation of Visium data. You can visualize the spatial distribution of UMI counts, gene counts, and identified spot clusters, as well as expression of individual genes of interest.

Spatial transcriptomics + single-cell RNA-seq: a powerful combination

When performing a spatial transcriptomics experiment, we recommend running a single-cell or single-nuclei RNA sequencing study in parallel. This is typically done using serial sections of the same tissue block. This provides several advantages.

Better identification of cell types. In a Visium experiment, each spot may capture up to 20 cells, and the spot RNA may be thought of as a “bulk RNAseq” population representing a small mixture of heterogeneous cells. Deconvoluting the spot transcriptome into separate cell types is typically done using a reference set of publicly available single cell RNAseq data for cell types that are believed to be in the tissue of interest. However, the quality and validity of cell type assignments is greatest when the single cell or single nuclei RNAseq data are obtained from the same Visium tissue sample. For example, Visium + single cell/single nuclei profiling of serial sections can capture rare or novel cell types that may be present in your tissue but missing from reference datasets.

Better detection of genes. Single-cell RNA-seq by definition captures individual cells and at greater sequencing depth per cell than is possible with spatial transcriptomes. Integration of single cell and spatial transcriptomes boosts average read, UMI and gene counts per cell providing more in-depth information from your spatial experiments.

Spatial Multiomics data integration

Simultaneous capture of multidimensional molecular information gives a much clearer picture of cell states. Computational methods serve to integrate multiomic datasets across experiments, conditions, and technologies.

At 3DG we primarily use Seurat, which is an R-based program with several tools for analyzing multi-modal single-cell data.

Seurat’s anchor-based integration method makes a series of cell-by-cell pairwise comparisons (from each dataset) and identifies “anchors” which represent cell pairs that share common features. Each anchor has a score based on the extent of overlap between the two cells, and then the data is integrated such that cells with a high anchor score come together while cells without a similar pair remain by themselves.

This allows information from one dataset to inform the other one, as in the case when using single-cell data to deconvolute Visium spots by cell type identity. This approach can be used to integrate single-cell RNA-seq, protein, chromatin accessibility, and spatial data.

How to work with us on a Visium experiment

Here at 3DG, we’ve performed Visium studies with multiple mouse and human tissue types, and can potentially work with any organism for which a reference genome sequence is available.

We work with both fresh-frozen and FFPE samples. In addition to tissue handling and library preparation, we provide computational analysis of spatial transcriptomics data and multiomic integration with single-cell or single-nuclei RNAseq as well as other datatypes such as protein or proteome and chromatin conformation (e.g., scATACseq).

We can work with you directly or through Science Exchange, a marketplace for research services.

What we provide

  • Tissue block sectioning and staining (H&E or immunofluorescence)
  • Tissue optimization (fresh-frozen tissues only) to determine the correct permeabilization time for your tissue
  • Tissue adhesion test (FFPE tissues only) to ensure FFPE blocks will adhere to the Visium Spatial Gene Expression Slide
  • Imaging of tissue sections (brightfield or fluorescence)
  • 10X Visium cDNA synthesis and library preparation
  • Sequencing, read mapping and alignment using SpaceRanger. 10X recommends sequencing 50K read pairs per spot covered by tissue (~250K read pairs for a capture area fully covered by tissue)
  • Bioinformatics on mapped reads including clustering, cell type annotation and differential gene expression
Source: 10X Genomics

What you provide

  • Tissue samples (fresh-frozen or FFPE) of sufficient quality for 10X Visium
    • To ensure the highest quality of your Visium libraries, we recommend samples have a DV200 >= 50% (where DV200 is the percentage of RNA fragments greater than 200 nucleotides). If your samples have a DV200 below 50%, we may be able to process them but note that library quality could be affected.
    • Ideally, the size of the tissue block is no greater than the Visium section area of 6.5mm x 6.5mm. If the tissue is larger, it will need to be scored to fit onto the capture area.
    • Fresh-frozen tissue:
      • 10X recommends snap-freezing tissue using an isopentane and liquid nitrogen bath (see their freezing guide here). Tissues frozen using other methods may be used, as long as sample quality is not affected.
      • We suggest assessing RNA quality by calculating the sample’s RNA Integrity Number (RIN). Samples with a RIN >= 7 will yield the best libraries.
    • FFPE tissue
      •  FFPE samples do not require a tissue optimization experiment, but do require a Tissue Adhesion Test to determine the risk of section detachment during Visium processing

What we deliver

  • 10X Visium Spaceranger output: web summary, feature-barcodes matrices, BAM files, automated secondary analyses, Loupe browser file (cloupe), and other output files
  • High-quality H&E (or immunofluorescence) images of your tissue sections
  • Raw sequencing files (FASTQ) and sequencing QC report
  • Computational support including clustering, cell type annotation and multiomic integration, if applicable
  • Remaining tissue samples will be returned to you

We look forward to working with you to get the best possible Visium libraries from your tissues. Contact us for more information, project inquiries and quotes. We can work with you directly or through Science Exchange, a marketplace for research services.