10x Visium HD Protocol 2.0 and 11 mm Chip: What Researchers Should Know
Inquiry
Figure 1. Visium HD workflow with Protocol 2.0 improvements highlighted. The probe-based chemistry and CytAssist transfer step are unchanged; sensitivity gains come from optimized hybridization, ligation, and wash conditions.
Visium HD from 10x Genomics brought single-cell-scale resolution to sequencing-based spatial transcriptomics when it launched in 2024. Instead of the 55-micron spots of standard Visium, Visium HD uses a continuous lawn of 2 × 2 μm barcoded squares, capturing whole-transcriptome data at near-single-cell resolution from FFPE and frozen tissues alike. Two recent upgrades extend what the platform can do. The first is Protocol 2.0, a revised experimental workflow that improves detection sensitivity—particularly for FFPE samples. The second is the 11 mm × 11 mm capture area, which roughly triples the tissue area a single slide can accommodate. Together, these changes affect sample preparation, experimental design, and the kinds of studies that Visium HD can support.
This article explains what each upgrade delivers, what stays the same, and how to factor the changes into study planning.
Protocol 2.0 and FFPE Sensitivity
Protocol 2.0 reworks the probe hybridization, ligation, and wash steps in the Visium HD workflow. The chemistry is the same probe-based whole-transcriptome panel, and the 2 μm barcoded array is unchanged. The gain comes from optimized reaction conditions that improve probe capture efficiency.
The headline result is an up to two-fold improvement in sensitivity compared with Protocol 1.0, measured by genes detected per 8-μm bin. The improvement is not uniform across sample types:
| Sample Type | Sensitivity Gain | Key Condition |
|---|---|---|
| High-quality FFPE | Up to ~2× | Fixation 6–24 hr, storage < 3 years |
| Fresh frozen (FF) | Modest, consistent | Standard freezing protocol |
| Fixed frozen (FxF) | Modest, consistent | Standard fixation and freezing |
| Low-quality / aged FFPE | Minimal to none | Low DV200 scores limit benefit |
- High-quality FFPE blocks show the largest gains. Tissues fixed within the recommended 6–24 hour window and stored for under three years benefit most.
- Fresh frozen (FF) and fixed frozen (FxF) samples show a modest but consistent improvement.
- The improvement scales with RNA quality—samples with low DV200 scores see less benefit.
This matters because FFPE is the most common archival format in clinical and translational research. Tumor biobanks, pathology archives, and multi-institutional cohorts are overwhelmingly FFPE. Protocol 2.0 makes Visium HD more viable for these sample types, where signal recovery has historically been the bottleneck.
A practical consequence: a 10 mm² tissue area that yielded 4,000 detectable genes per 8-μm bin under Protocol 1.0 might yield roughly 5,500–8,000 genes under Protocol 2.0, depending on sample quality. For discovery-driven studies, the additional gene coverage means more statistical power for detecting spatially variable genes, and more complete pathway coverage in downstream enrichment analyses.
The 11 mm Capture Area
The standard Visium HD capture area is 6.5 mm × 6.5 mm (42 mm²). The new 11 mm × 11 mm format expands this to 121 mm²—roughly a three-fold increase.
| Feature | 6.5 mm Visium HD | 11 mm Visium HD |
|---|---|---|
| Capture area | 6.5 × 6.5 mm (42 mm²) | 11 × 11 mm (121 mm²) |
| Barcoded feature size | 2 μm continuous | 2 μm continuous |
| Default analysis bin | 8 μm | 8 μm |
| Capture areas per slide | 2 (A1, D1) | 2 (A, B) |
| CytAssist firmware | v2.0+ | v2.4.0+ |
| Tissue coverage | Standard sections (~1 cm²) | Whole-organ sections, TMAs |
Figure 2. Scale comparison of the 6.5 mm and 11 mm Visium HD capture areas. The 11 mm format provides approximately three times the capture area, enabling whole-organ sections and multi-sample TMAs within a single capture area.
The larger format changes what is possible in a single capture area:
- Whole-organ sections from mouse and small-organ human samples can be profiled without tiling or stitching, preserving continuous spatial context across the entire tissue.
- Tissue microarray (TMA) studies can fit more cores per capture area, reducing the per-sample cost and improving throughput for cohort-scale spatial profiling.
- Multi-region sampling of large human surgical specimens—for example, capturing both the tumor core and the invasive margin in a single capture area—becomes feasible without sacrificing spatial coverage.
- Tissue co-embedding allows multiple smaller samples (e.g., mouse organs from a single experiment) to share one capture area, improving cost efficiency for multi-tissue studies.
The 11 mm chip requires CytAssist firmware v2.4.0 or later. For labs with instruments running older firmware, a firmware update is needed before using the larger slides. The 6.5 mm slides continue to work with the updated firmware, so existing workflows are not disrupted.
What Space Ranger 4.1 Adds
Alongside the hardware and protocol upgrades, Space Ranger v4.1 introduces automatic cell type annotation. This feature is available for Visium HD libraries and can be run either as part of the standard spaceranger count pipeline or as a standalone spaceranger annotate command.
The annotation system uses two model types:
- Cloud-based models (co-developed with the Broad Institute's Cellarium AI Lab): support human and mouse samples, require a 10x Cloud account.
- Pan-Human Azimuth model (developed by the Satija Lab / HuBMAP): supports human samples only, runs locally without cloud access.
Annotation output is stored alongside standard Space Ranger outputs and can be viewed in Loupe Browser v9.1+. For labs that already perform cell-type annotation downstream using tools such as SingleR, CellTypist, or reference-based mapping in Seurat, the built-in annotation provides a fast first-pass result that can reduce hands-on analysis time.
Space Ranger 4.1 also adds full support for the 11 mm capture area format.
Figure 3. Space Ranger 4.1 analysis pipeline with the new automatic cell-type annotation module. Cell-type labels are stored alongside standard outputs and can be inspected in Loupe Browser v9.1+.
What Stays the Same
Amid the upgrades, several fundamentals are unchanged:
- Probe-based chemistry. Visium HD still uses paired probe hybridization and ligation, targeting approximately 18,000 human or 20,000 mouse protein-coding genes. The probe panel is unchanged between Protocol 1.0 and 2.0.
- CytAssist workflow. Tissue sections are mounted on standard glass slides, stained (H&E or immunofluorescence), imaged, and then transferred to the Visium HD slide via the CytAssist instrument. This step controls spatial fidelity by limiting lateral diffusion of probes during transfer.
- 2 μm continuous barcoded lawn. The underlying capture surface is identical. Spatial resolution at the hardware level is unchanged.
- Analysis binning. The standard analysis pipeline still produces data at 2 μm, 8 μm, and 16 μm bin resolutions. The 8 μm bin remains the recommended default for most analyses.
- Sample compatibility. FFPE, fresh frozen, and fixed frozen samples are all supported, as before. The improvement in FFPE sensitivity under Protocol 2.0 does not change which tissue types are compatible.
Backward compatibility is a deliberate part of this upgrade: researchers with ongoing studies using 6.5 mm slides and Protocol 1.0 can continue without disruption, and data from both protocols can be analyzed together in Space Ranger 4.1.
Planning a Visium HD Experiment
For researchers planning a new Visium HD study, the upgrades affect several practical decisions.
Sample preparation. If working with FFPE tissue, prioritize blocks with fixation times in the 6–24 hour range and storage under three years. Request a DV200 measurement if feasible—this RNA quality metric is a better predictor of Visium HD performance than the more common DIN score, because the probe-based assay depends on moderately sized RNA fragments rather than full-length transcripts.
Chip selection. Choose the 11 mm format when tissue sections exceed roughly 8 mm in either dimension, when running TMA studies, or when capturing multiple tissue regions in a single capture area is scientifically valuable. For standard single-region sections under 6 mm, the 6.5 mm format remains appropriate and may reduce sequencing cost.
Data analysis. Factor Space Ranger 4.1 into the analysis pipeline timeline. The built-in cell-type annotation can accelerate initial data exploration, but may need to be supplemented with project-specific reference datasets for publication-quality annotation. For human studies, the local Azimuth model avoids cloud dependency and is a practical default.
Sequencing depth. The standard recommendation of approximately 275 million read pairs per fully covered 6.5 mm capture area scales with capture area. The 11 mm slide, when fully covered, requires proportionally more sequencing—plan for roughly 800 million read pairs for full coverage at standard depth. Under-coverage is possible for pilot studies or when only a portion of the capture area is occupied.
Where Visium HD Fits Now
The spatial transcriptomics platform landscape has expanded rapidly. A 2025 systematic benchmarking study compared Visium HD with other subcellular-resolution platforms (Stereo-seq, Xenium 5K, CosMx 6K) across human tumor samples and found that Visium HD's strength lies in tissue-region-level analysis with near-whole-transcriptome coverage. Imaging-based platforms generally offered cleaner single-cell boundary definition, while sequencing-based platforms including Visium HD provided broader gene coverage that supports discovery-oriented analysis. For researchers weighing platform options, a guide to spatial transcriptomics platforms covers the trade-offs between sequencing-based and imaging-based approaches in more detail.
Protocol 2.0 strengthens this position. The sensitivity gain for FFPE brings Visium HD closer to the detection efficiency of fresh-frozen protocols for the sample types that dominate translational research. And the 11 mm chip removes a practical ceiling—whole-slide tissue imaging, once limited to imaging-based platforms with smaller fields of view, is now accessible within a sequencing-based whole-transcriptome framework.
As the field moves toward true multi-modal spatial profiling (see the preview of Visium HD Pro with 206-plex protein co-detection), Protocol 2.0 and the 11 mm chip lay groundwork for studies that demand both transcriptome breadth and large-area tissue coverage from FFPE samples.
Figure 4. Conceptual illustration of sensitivity improvement from Protocol 1.0 to Protocol 2.0. The gain is most pronounced for high-quality FFPE samples, with more genes detected per 8-μm analysis bin.
For research teams evaluating whether Visium HD fits their project, spatial transcriptomics services can provide sample feasibility review, from tissue QC through platform selection. Researchers working with archival FFPE cohorts may also benefit from FFPE spatial transcriptomics services specifically optimized for the sample type that Protocol 2.0 benefits most.
FAQs
Is Protocol 2.0 backward compatible with existing Visium HD 6.5 mm slides?
Yes. Protocol 2.0 is fully backward compatible with the standard 6.5 mm Visium HD slides. Researchers can upgrade their experimental protocol without replacing their existing slide inventory. Data from Protocol 1.0 and Protocol 2.0 can be analyzed together in Space Ranger 4.1.
Does the 11 mm chip require new instrumentation or hardware?
The 11 mm chip uses the same CytAssist instrument and workflow as the 6.5 mm format. The only requirement is a firmware update to CytAssist v2.4.0 or later. No new hardware purchase is needed, and the updated firmware continues to support the 6.5 mm slides.
How much additional sequencing do I need for the 11 mm chip?
Sequencing depth scales with capture area. The standard recommendation of approximately 275 million read pairs per fully covered 6.5 mm capture area translates to roughly 800 million read pairs for a fully covered 11 mm capture area. If only a portion of the capture area is occupied by tissue, proportionally fewer reads are needed. Under-coverage is a practical option for pilot studies.
Can I combine data from Protocol 1.0 and Protocol 2.0 in the same analysis?
Yes. Space Ranger 4.1 processes data from both protocols, and the underlying probe panel, barcoded array, and analysis binning are identical. The sensitivity difference between protocols should be treated as a batch variable in downstream analysis. For studies that include samples processed under both protocols, include protocol version as a covariate in statistical models, particularly for differential expression testing where sensitivity differences could confound results.
This content is intended for research use only. The methods, tools, and platforms described are not intended for clinical diagnosis, treatment decisions, or individual health assessment.
References
- Oliveira MF, Romero JP, Chung M, Williams S, Gottscho AD, Gupta A, Pilipauskas S, Mohabbat S, Raman N, Sukovich D, Patterson D, Visium HD Development Team, Taylor SEB. High-definition spatial transcriptomic profiling of immune cell populations in colorectal cancer. Nature Genetics. 2025;57(6):1512-1523.
- Ren P, Zhang R, Wang Y, Zhang P, Luo C, Wang S, Li X, Zhang Z, Zhao Y, He Y, Zhang H, Li Y, Gao Z, Zhang X, Zhao Y, Liu Z, Meng Y, Zhang Z, Zeng Z. Systematic benchmarking of high-throughput subcellular spatial transcriptomics platforms across human tumors. Nature Communications. 2025;16:9232.
- 10x Genomics. Visium HD Protocol 2.0 and 11 mm Capture Area. Product documentation and official benchmark data. 2026. Available at: https://www.10xgenomics.com/platforms/visium
- 10x Genomics. Space Ranger 4.1 Release Notes. 2025.
- Lim HJ, Wang Y, Buzdin A, Li X. A practical guide for choosing an optimal spatial transcriptomics technology from seven major commercially available options. BMC Genomics. 2025;26:47.
- Gulati GS, D'Silva JP, Liu Y, Wang L, Newman AM. Profiling cell identity and tissue architecture with single-cell and spatial transcriptomics. Nature Reviews Molecular Cell Biology. 2025;26(1):11-31.
