Human Breast Dataset Explorer

This is a preview of Xenium In Situ data from a development version of Xenium.

Go to www.10xgenomics.com/resources/datasets to view new data from the improved production version of Xenium Analyzer.

Key Metrics

Species and tissue
Human breast
Preservation method
FFPE
Off-the-shelf panel used
Xenium Human Breast Gene Expression Panel (280 genes)
Number of custom genes
33
Tissue area
0.4 cm²
Number of cells detected*
167,780
Total transcripts detected*
34,442,716
Median transcripts per cell*
166

*Metrics may differ slightly from preprint because this dataset was processed using a newer version of the analysis pipeline

Explore cell type heterogeneity with Xenium and verify with H&E staining on the same section.

Different tissue macrostructures are revealed by the spatial organization of cell types and gene features, capturing vast cellular and molecular heterogeneity within tumor regions

Xenium identifies major cell types in the human breast. As shown in two examples, spatial information is necessary to study tumor subtype heterogeneity and the microenvironment surrounding tumor epithelial cells. Interpretation and validation of rich transcript information is assisted by H&E staining acquired post-Xenium.

Cells colored by expression profiles in tSNE space and mapped back to original tissue space — Cells colored by expression profiles in t-SNE space and mapped back to original tissue space (top). H&E image of the same section. Scale bar 1 mm (bottom).

Cells colored by expression profiles in t-SNE space and mapped back to original tissue space (top). H&E image of the same section. Scale bar 1 mm (bottom).

Atypical ductal hyperplasia. Myoepithelial cells (green) are present in healthy breast lining the basal cell layer of epithelium. In hyperplasia, some cells exhibit expression profiles more similar to cells typically found in ductal carcinoma in situ (DCIS).

DCIS (left) in proximity to benign duct ectasia (right) — DCIS (green arrow) in proximity to benign duct ectasia (orange arrow). In duct ectasia, macrophages are detected within the thickened duct wall, surrounded by inflammatory cells. Scale bar 200 μm.

DAPI-based cell segmentation

Xenium uses an in-house trained model to segment nuclei from the DAPI image. Nuclei of irregular shapes are segmented well using Xenium analysis software. Transcripts are assigned based on expansion from nuclei.

DAPI image with nuclear outlines. Scale bar: 50 μm

DAPI image with nuclear (green) and cellular (yellow) outlines. Scale bar: 50 μm

Xenium output can be analyzed in commonly used single cell data analysis software to annotate clusters and/or identify differentially expressed genes.

Top 10 differentially expressed genes in example cell types

Find triple-positive rare cells

In what is classified as ERBB2(HER2)-positive, ESR1(ER)-positive and PGR(PR)-negative breast tumor, Xenium finds rare triple-positive cells that express all three hormonal receptors.