The Xenium platform uses targeted panels to detect gene expression at subcellular resolution. The pre-designed panels were designed to cover the major cell types for specific tissues. They consist of genes selected from single cell atlas data, research area experts, and literature searches.
Consult the Xenium In Situ Gene Expression - Probe Hybridization, Ligation & Amplification User Guide (CG000582) to execute the workflow.
Table of contents
- Pre-designed panel information
- Custom panel information
- Panel metadata
- Probe genomic target sequences and coordinates
Pre-designed panel information
The table below provides the following information for available pre-designed Xenium gene expression panels:
- Panel name and the number of targeted genes
- Downloadable panel metadata with gene name, transcript ID, number of probe sets, number of codewords for each target, and cell type annotation in CSV format (see below for details).
gene_panelJSON file for each panel (format described here).
- Panel probe sequences in FASTA format and the genomic coordinates of the panel probe sequences in 12-column detailed BED format (see below for details).
|Panel metadata and JSON
|Probe genomic target sequences and coordinates
|Xenium Human Breast Gene Expression Panel
|Xenium Mouse Brain Gene Expression Panel
|Xenium Human Brain Gene Expression Panel
|Xenium Human Lung Gene Expression Panel
|Xenium Human Multi-Tissue and Cancer Panel
|Xenium Mouse Tissue Atlassing Panel
|Xenium Human Colon Gene Expression Panel
|Xenium Human Skin Gene Expression Panel
Panel negative controls
Each pre-designed panel contains 40 negative control codewords and 20 negative control probe sets (except the Xenium Mouse Brain Gene Expression Panel, which has 27).
For negative control probe sets, multiple probes are designed for each non-biological target sequence. It follows the same design process as for real target sequences, with a goal of reaching eight probe sets per target (for more detail, read the Getting Started with Xenium Panel Design page). An add-on custom panel will have the same number of negative control probe sets as the pre-designed panel it is based upon.
Decoding and negative controls are described on the Xenium Onboard Analysis algorithms page.
Custom panel information
Please read the Getting Started with Xenium Panel Design page for information about supported add-on custom and standalone custom gene panel options.
For each of the pre-designed panels, a metadata file is provided on this page (CSV format). This file contains additional panel information, which may be helpful for designing your experiment or custom add-on genes.
|The gene identifier targeted by the probe.
|The Ensembl gene identifier targeted by the probe.
|Number of Probe sets
|The number of probe sets targeting each isoform of the gene.
|The number of codewords used to decode each gene.
|The 10x Genomics cell type annotation given to each gene based on single cell atlas data, research area experts, and literature searches.
Probe genomic target sequences and coordinates
The FASTA file contains the panel probe sequences in FASTA format, as represented on the probes and reverse-complement to the targeted transcript in FASTA format. The sequence ID consists of the transcript ID, gene name, and a unique identifying string, e.g.
The 12 column BED-formatted file contains the sequences and genomic coordinates of the probes. The fourth column corresponds to the sequence ID of the FASTA file. The underlying genomic references are GRCh38 for human and GRCm38 for mouse.
This file may be used to visualize the probe locations with genome browsers like IGV (Integrated Genomics Viewer) and the UCSC Genome Browser, or to intersect the probe locations with other genomic features of interest using tools like bedtools.
The BED file information is useful for determining whether probes span exon-exon junctions. For example, probe
ENSG00000134440|NARS|1f419e5, has the following sequence:
And its BED record shows the coordinates in the top row below:
chr18 57613656 57615656 ENSG00000134440|NARS|1f419e5 0 + 57613656 57615656 0 2 24,16 0,1984
Loading the BED file in IGV shows this probe spans two exons:
By zooming in, we see the probe sequence as represented in the FASTA file is the reverse-complement of the transcript sequence: