Feb 25, 2026
BIDMC TMC / STU - Methodological compendium for 10x Genomics Xenium spatial transcriptomics
- Shuoshuo Wang1,2,3,
- Antonella Arruda de Amaral1,2,
- aploumak 1,3,
- Sheethal Umesh Nagalakshmi1,2,3,
- Ioannis Vlachos1,2,3
- 1Beth Israel Deaconess Medical Center;
- 2Broad Institute of MIT and Harvard;
- 3Harvard Medical School
- Human BioMolecular Atlas Program (HuBMAP) Method Development CommunityTech. support email: [email protected]
Protocol Citation: Shuoshuo Wang, Antonella Arruda de Amaral, aploumak , Sheethal Umesh Nagalakshmi, Ioannis Vlachos 2026. BIDMC TMC / STU - Methodological compendium for 10x Genomics Xenium spatial transcriptomics . protocols.io https://dx.doi.org/10.17504/protocols.io.5jyl8qz5rl2w/v1
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: May 15, 2025
Last Modified: February 25, 2026
Protocol Integer ID: 218376
Keywords: Xenium, Tissue mapping, Spatial, Spatial transcriptomics, smFISH, single molecule FISH, Imaging, Tissue preparation, Sample preparation, Xenium, Spatial Transcriptomics, In Situ Hybridization, smFISH, Lymphatic Vasculature, Tissue Mapping, Human Reference Atlas, HuBMAP, 10x genomics xenium, 10x genomics xenium platform, methodological compendium for 10x genomics xenium, reproducible workflow for in situ spatial transcriptomic, situ spatial transcriptomic, spatial gene expression data into atlas, hubmap tissue mapping center, spatial transcriptomic, spatial gene expression data, scale tissue mapping initiative, leveraging xenium, tissue architecture, subcellular transcript detection in both ffpe, subcellular transcript detection, resolution molecular map, reconstruction of tissue architecture, lymphatic vasculature within human cervix, localization of transcriptional program, human cervix, including rna integrity, applicable across tissue type, lymphatic network, interoperable spatial biology, rna integrity, co
Funders Acknowledgements:
National Heart Lung and Blood Institute
Grant ID: U54HL165440
Disclaimer
DISCLAIMER – FOR INFORMATIONAL PURPOSES ONLY; USE AT YOUR OWN RISK
The protocol content here is for informational purposes only and does not constitute legal, medical, clinical, or safety advice, or otherwise; content added to protocols.io is not peer reviewed and may not have undergone a formal approval of any kind. Information presented in this protocol should not substitute for independent professional judgment, advice, diagnosis, or treatment. Any action you take or refrain from taking using or relying upon the information presented here is strictly at your own risk. You agree that neither the Company nor any of the authors, contributors, administrators, or anyone else associated with protocols.io, can be held responsible for your use of the information contained in or linked to this protocol or any of our Sites/Apps and Services.
Abstract
We present a standardized and reproducible workflow for in situ spatial transcriptomics using the 10x Genomics Xenium platform, developed and operationalized by the HuBMAP Tissue Mapping Center at Beth Israel Deaconess Medical Center. Although broadly applicable across tissue types, the protocol is informed by extensive practical experience in mapping lymphatic vasculature within human cervix, heart, lung, and skin. The workflow details optimized procedures for tissue preparation, assay configuration, and imaging, leveraging Xenium’s padlock-probe chemistry and rolling circle amplification to achieve subcellular transcript detection in both FFPE and fresh-frozen specimens. Emphasis is placed on pre-analytical variables including RNA integrity, fixation compatibility, and imaging calibration, to ensure signal fidelity, spatial accuracy, and cross-sample reproducibility. Beyond experimental execution, the protocol provides an adaptable framework for integration of spatial gene expression data into atlas-scale and disease-focused investigations. The resulting high-resolution molecular maps enable robust downstream analyses, including cell-type annotation, reconstruction of tissue architecture, and localization of transcriptional programs within rare or structurally complex compartments such as lymphatic networks. By aligning with community standards and multi-platform harmonization efforts, this protocol facilitates reproducible, interoperable spatial biology within large-scale tissue mapping initiatives.
Materials
Core Equipment and Hardware
Xenium Analyzer (10x Genomics): An automated system for fluidics management and wide-field epifluorescence imaging;
Xenium Slides (10x Genomics);
Xenium Cassette, Lid, and Thermocycler Adaptor: Essential components for slide assembly, reagent delivery, and thermal cycling;
Microtome or Cryostat: For sectioning FFPE or Fresh Frozen tissue blocks, respectively;
Slide Warming Tray or Oven: Capable of maintaining a precise temperature of for tissue baking.
Vacuum Desiccator and Desiccant: Used for overnight moisture elimination post-baking.
Thermal Cycler: Required for controlled high-temperature incubations during decrosslinking and hybridization.
Tissues and Processing Reagents
Tissue Blocks: Either Fresh Frozen (FF) or Formalin-Fixed, Paraffin-Embedded (FFPE).
Deparaffinization and Rehydration Solvents: Includes Xylene, 100% Ethanol, 96% Ethanol, 70% Ethanol, and Nuclease-free Water.
Decrosslinking Agents: Xenium Enhancer and 0.5 M Urea for reversing formaldehyde-induced methylene bridges.
Xenium Kit Reagents and Assay Components
Xenium Decoding Reagents and Consumables: Comprehensive reagents for the in situ hybridization and decoding process.
Targeted Padlock Probes ("Panel"): Specific circularizable DNA probes for mRNA detection.
Safety Equipment: Gloves, lab coat, and eye protection.
Staining Jars or Containers: For processing slides through various gradients and washes.
Buffers: 1X PBS and TE Buffer for probe resuspension.
Troubleshooting
The Xenium platform
In a nutshell, the 10x Genomics Xenium platform, is an highly multiplexed in situ spatial analysis system that utilizes padlock-probe-based sequence detection coupled with rolling circle amplification to achieve true subcellular and near-cellular resolution. Engineered to simultaneously detect hundreds to thousands of distinct RNA transcripts, the platform exhibits robust compatibility with both fresh frozen (FF) and formalin-fixed, paraffin-embedded (FFPE) tissue samples.
Goals:
Detailed characterization of cellular heterogeneity within complex tissues.
Creation of spatial reference maps for defined anatomical structures.
Understanding of how cellular organization and gene expression contribute to tissue function in health and disease.
This protocol provides a standardized approach to generate high-quality Xenium data, contributing to the goals of initiatives like HubMAP by providing detailed molecular and spatial information essential for building a comprehensive atlas of the human body at the cellular level.
Pre-analytical quality control and tissue preparation guidelines
Gentle tissue handling is imperative to avoid structural damage such as coagulative necrosis stemming from surgical electrocautery or oxidized hemorrhages resulting from blunt surgical trauma.
Prolonged ischemia and extended post-mortem intervals (PMI) exponentially decrease measurable transcript density. Therefore, depending on sample types, the specific protocol must stipulate a maximum allowable interval, typically not exceeding four hours, between physical tissue resection and submersion in fixative. In BIDMC-TMC internal practice this time is kept between 1 hour maximum.
Critically, the physical orientation of the tissue during embedding must be meticulously recorded to ensure proper spatiality for downstream functional region selection and analysis within the Human Reference Atlas.
HuBMAP-specific disambiguation:
"Sample Block": the macroscopic, paraffin-embedded tissue matrix;
"Sample Section": the discrete 5-micrometer microtome-cut slice placed onto the specialized Xenium slide;
"Assay": the comprehensive Xenium in situ hybridization and decoding process.
Sample preparation for FFPE
3h 40m
If starting with FFPE blocks that require rehydration prior to sectioning, rehydrate the block face by placing it on a chilled surface (e.g., ice bath or cold plate) for 10-30 minutes. This step may vary based on tissue type and specific laboratory practices.
Once successfully placed, the slides must be dried upright at room temperature until the tissue appears opaque. Subsequently, the protocol mandates baking the slides in an oven or on a thermal slide warmer at exactly 42 °C for a duration of 03:00:00 . This prolonged baking step promotes covalent adhesion between the tissue proteins and the proprietary slide coating.
Finally, the baked slides must be stored overnight in a vacuum desiccator containing fresh desiccant to completely eliminate residual moisture, preventing downstream tissue detachment during the aggressive heated wash steps.
3h
Deparaffinization and Tissue Rehydration
After Paraffin removal, the descending ethanol gradient systematically and gently rehydrates the biological tissue, preparing the cellular matrix for the aqueous enzymatic reactions that constitute the core of the assay.
Section Fresh Frozen tissue onto Xenium Slides (based on Demonstrated Protocol CG000579)
This section details the critical steps for preparing tissue sections on Xenium slides. Adherence to these guidelines is crucial for optimal assay performance.
The Xenium slide itself features a highly specific, proprietary optical architecture. It contains an active, imageable scan area measuring exactly 10.45 mm x 22.45mm, surrounded by a dense frame of fiducial markers.
While it is functionally permissible for excess, blank paraffin wax to overlap the fiducial markers, the actual dense biological tissue must absolutely not obscure these markers. The Xenium Analyzer relies entirely on these fiducials for the complex sub-micron spatial orientation and multi-cycle image registration required during the decoding phase.
discontinued:
Xenium Cassette Assembly and Chemical Decrosslinking (Based on User Guide, CG000623)
3h 40m
Following successful rehydration, the slide is carefully assembled into the proprietary Xenium Cassette. Xenium Quick Reference Cards for Slide Cassette Assembly.
Operators must apply even, bilateral downward pressure until the integrated clips audibly click into place, securing the leakproof elastomeric gasket. Any visual gap between the upper and lower cassette halves indicates a structural failure that will inevitably result in reagent leakage, leading to tissue desiccation and catastrophic assay failure.
Once the cassette is securely assembled and loaded with 1X PBS, the tissue undergoes crucial chemical decrosslinking to revert dense, covalent methylene bridges between cellular proteins and nucleic acids by fixation, rendering the target RNA physically inaccessible to the bulky hybridization probes.
The cassette is then sealed with a Xenium Cassette Lid and placed onto a Xenium Thermocycler Adaptor. The thermal cycler is programmed with a heated lid set to 80 °C , and the slide is incubated at 80 °C for exactly 00:30:00 , followed by an automated re-equilibration step at 22 °C for 00:10:00 . This highly controlled, high-temperature enzymatic and chemical incubation breaks the covalent methylene bridges, effectively unmasking the transcriptomic targets without inducing thermal degradation of the RNA.
40m
Probe Hybridization, Ligation & Amplification (based on User Guide CG000582)
4h 54m
The foundational chemistry of the Xenium platform relies on highly specific circularizable padlock probes. Each individual DNA probe consists of two distinct target-homology regions that are designed to bind to adjacent sequences on the specific target mRNA. These homology regions are separated by a linking sequence containing a unique, gene-specific barcode and a universal primer binding site.
Depending on the exact experimental design of the HuBMAP study, this mix may utilize entirely standalone custom probe panels. Custom probes, which are delivered in a lyophilized state, must first be carefully resuspended in TE Buffer (8 ) and allowed to equilibrate. Immediately prior to usage, the complex probe mixtures must be preheated at 95 °C for 00:02:00 to melt intermolecular secondary structures and dimers, followed immediately by snap-cooling on ice for 00:01:00 to lock the probes in a monomeric state.
The final Probe Hybridization Mix is prepared by combining Probe Hybridization Buffer (PN-2000390) with the snap-cooled probes and additional TE Buffer to yield a final reaction volume of 500 µL per slide. This mix is gently dispensed along the side of the cassette well to avoid disturbing the decrosslinked tissue. The sealed reaction is then incubated Overnight (spanning 16 to 24 hours) in a thermal cycler maintained at a constant 50 °C . The thermodynamic stringency established by this elevated temperature ensures that only padlock probes exhibiting perfect Watson-Crick base-pairing complementarity to the target mRNA will stably hybridize.
3m
Probe Ligation and Rolling Circle Amplification (RCA)
Following the overnight hybridization, the vast majority of probes will remain unbound in solution. These must be aggressively removed via a stringent wash utilizing Post Hybridization Wash Buffer (PN-2000395), incubated at 37 °C for 00:30:00 . The protocol then advances to the critical enzymatic ligation phase.
The specificity of the padlock probe architecture lies in the requirement for physical spatial proximity. Only when both the 3' and 5' homology regions of a single probe bind adjacent to one another on the target RNA strand can trigger the ligation sealing the nick, transforming the linear probe into a covalently closed circular DNA molecule. The protocol specifies the preparation of a Ligation Mix comprising Ligation Buffer (PN-2000391), Ligation Enzyme A (PN-2000397), and Ligation Enzyme B (PN-2000398). 500 µL of this mixture is applied to the tissue and incubated at 37 °C for 02:00:00 .
Once successfully circularized, the probes act as endless templates for Rolling Circle Amplification (RCA). A highly processive, strand-displacing polymerase (derived from Phi29) is added via the Amplification Master Mix, consisting of Amplification Mix (PN-2000392) and Amplification Enzyme (PN-2000399). Incubated at a lower temperature of 30 °C for 02:00:00 , the polymerase continuously travels around the circularized DNA probe, sequentially generating a massive, long, single-stranded DNA concatemer. This resulting structure often termed a DNA "nanoball" contains hundreds of tandem repeating copies of the gene-specific barcode sequence. This massive, physically localized target amplification is the core mechanism that provides the Xenium assay with its exceptional signal-to-noise ratio, enabling robust single-molecule detection against high background noise.
4h 30m
Autofluorescence Quenching and Nuclei Staining
Human tissues, particularly complex clinical samples often exhibit high levels of broad-spectrum intrinsic autofluorescence. This phenomenon is primarily driven by the accumulation of lipofuscin, dense crosslinked elastin fibers, and heavily oxidized red blood cells. If left untreated, this overwhelming background autofluorescence completely obscures the specific fluorescent signals generated by the decoding probes during the instrument run.
The protocols.io document must therefore prescribe the rigorous Autofluorescence Quenching steps in exacting detail. The tissue is initially treated with a combination of 1X PBS washes, followed by an incubation with Diluted Reducing Agent B (PN-2000087) for 00:10:00 to chemically neutralize specific fluorophores. The tissue is subsequently dehydrated via a rapid series of ethanol washes and treated with a specialized Autofluorescence Solution (comprising Autofluorescence Mix PN-2000753 diluted in 100% Ethanol) for 00:10:00 . Crucially, this step must be performed strictly in the dark to prevent photobleaching of the quenching agents.
20m
Summary:
| A | B | C | D | |
| Assay Stage | Primary Reagents Utilized | Defined Incubation Parameters | Core Biochemical Function | |
| Decrosslinking | Xenium Enhancer, 0.5 M Urea | 80∘C for 30 minutes | Hydrolyzes formaldehyde-induced methylene crosslinks. | |
| Hybridization | Targeted Padlock Probes | 50∘C for 16-24 hours | Drives sequence-specific binding to unmasked mRNA. | |
| Ligation | Ligase Enzymes A & B | 37∘C for 2 hours | Covalently circularizes perfectly hybridized, adjacent probes. | |
| Amplification | Strand-Displacing Polymerase | 30∘C for 2 hours | Synthesizes massive RCA DNA nanoballs for signal boosting. | |
| Quenching | Reducing Agent B, AutoFL Mix | Room Temp for 20 minutes | Eliminates endogenous tissue fluorescence (lipofuscin/elastin). |
Probe Hybridization, Ligation & Amplification (based on User Guide CG000582)
4h 54m
Segmentation:
Spatial segmentation relied heavily on DAPI-based nuclear staining coupled with geometric expansion algorithms. While computationally inexpensive and experimentally straightforward, this unimodal approach may systematically misassigns transcripts, particularly in densely packed epithelial tissues or among morphologically complex cells such as neurons, astrocytes, and specialized immune subsets. To achieve true single-cell and even subcellular resolution, Xenium has pivoted toward multimodal segmentation strategies. These strategies integrate nuclear markers, cytoplasmic interior stains, synthetic lipid or glycan dyes, and highly specific protein-based immunofluorescence to generate high-fidelity boundary masks that reflect true biological morphology.
The integration of additional molecular markers introduces a significant "hassle factor" into experimental workflows. This hassle factor encompasses the requirement for extensive assay optimization, the management of prolonged incubation times, the mitigation of background autofluorescence, and the delicate balancing act of applying tissue permeabilization agents without jeopardizing the structural integrity of the tissue or exacerbating the degradation of the target RNA. The pursuit of the perfect segmentation boundary often directly conflicts with the foundational requirement of spatial transcriptomics: the preservation and sensitive detection of endogenous mRNA.
Xenium relies on a dedicated nuclear stain (DAPI) and cell boundary stain (e.g., ATP1A1/CD45 cocktail) processed via a deep-learning segmentation algorithm to assign transcripts to cells.
Nuclear Demarcation: The Baseline Modality
The foundational step of nearly all spatial transcriptomic segmentation pipelines involves the identification of the cell nucleus. The assumption underlying this unimodal approach is that the nucleus represents a universal, easily targetable cellular centroid from which the total cytoplasmic boundaries can be mathematically inferred. While indispensable as a starting coordinate, the nucleus alone is fundamentally insufficient for capturing the morphological reality of a cell.
Xenium protocol uses Nuclei Staining Buffer (PN-2000762), which contains DAPI, for 00:01:00 in the dark. This rapid step brightly stains the dense chromatin within the cellular nuclei. The resulting DAPI signal provides the foundational morphological reference map utilized by the Xenium Analyzer's onboard neural network to establish rudimentary 2D cell segmentation boundaries.
In the 10x Genomics Xenium multimodal segmentation workflow, the 18S rRNA stain is deployed as a critical interior morphological marker. The Xenium onboard algorithm utilizes a "nuclear expansion to interior edge" method. Starting from the DAPI-defined nucleus, the algorithm mathematically expands the cellular boundary outward until it detects the terminal edge of the 18S rRNA fluorescent signal. This biologically informed boundary tightly correlates with true cellular volume, effectively mitigating the arbitrary and artifact-prone nature of fixed-distance geometric expansion.
In addition, an epithelial and boundary marker Sodium-Potassium ATPase (NaK ATPase / ATP1A1) are heavily utilized as the gold standard for boundary definition. NaK ATPase, an integral membrane pump, is an exceptionally reliable and dense marker for the plasma membrane of tightly packed epithelial cells. These boundaries serve as a "ground-truth" reference.
Because lymphocytes and macrophages lack epithelial adhesion molecules, relying solely on epithelial markers results in the failure to segment the immune compartment. Therefore, the inclusion of pan-leukocyte markers like CD45 is beneficial in multimodal cocktails.
Advanced Algorithmic Cell Segmentation Strategies
Two principal community-standard approaches: Cellpose and Baysor. Both workflows produce a structured feature-by-cell count matrix in Matrix Market (MTX) format. This standardized output can be imported into downstream analytical frameworks such as Seurat (R) or Scanpy (Python) for dimensionality reduction, unsupervised clustering, UMAP embedding, and reference-based cell type annotation.
Cellpose (3D DAPI-Based Segmentation):
Cellpose employs a deep convolutional neural network to delineate discrete nuclear boundaries in volumetric space. Because Cellpose does not natively process multi-resolution pyramidal TIFFs, the protocol must instruct extraction of a single defined resolution level (e.g., Level 2 at 0.85 µm/pixel) from the morphology.ome.tif pyramid prior to analysis. All critical CLI parameters must be disclosed, including manual specification of --diameter (e.g., 7–10 µm for murine cortical nuclei, converted to pixel units based on the selected resolution) and activation of the --do_3D flag. Known computational constraints must also be addressed: for image arrays exceeding 4 GiB, users should resolve the NumPy OverflowError during serialization by forcing pickle.HIGHEST_PROTOCOL within the relevant format.py routine. Following 3D nuclear segmentation, transcript coordinates from transcripts.parquet are mapped to segmented nuclei, with a clearly defined nuclear expansion heuristic (e.g., 10 µm isotropic radial dilation) applied to capture perinuclear cytoplasmic transcripts.
Baysor (Transcript-Composition-Based Segmentation):
Alternatively, Baysor (Julia-based) infers cell boundaries directly from spatial RNA density and transcriptional composition rather than relying exclusively on DAPI morphology. The protocol must define preprocessing steps (e.g., Python/pandas filtering of transcripts with QV < 20 and reassignment of extracellular cell IDs from −1 to 0 to prevent runtime failure). Including low-quality transcripts significantly degrades Baysor’s composition-based clustering.
Reproducibility requires explicit reporting of execution parameters, including the minimum transcript threshold (-m 15) and prior segmentation weighting (e.g., --prior-segmentation-confidence 0.5) to incorporate the Xenium onboard mask as a probabilistic prior.
Both workflows produce a structured feature-by-cell count matrix in Matrix Market (MTX) format. This standardized output can be imported into downstream analytical frameworks such as Seurat (R) or Scanpy (Python) for dimensionality reduction, unsupervised clustering, UMAP embedding, and reference-based cell type annotation.
1m
Sample storage and transport
Following completion of the final wash series, implement validated safe stopping points to minimize temporal degradation of the highly labile RCA amplicons.
For short-term interruptions (≤1 week), retain the sealed cassette at 4 °C in the dark, fully immersed in PBS-T, refreshing the buffer every 3–4 days to mitigate microbial growth.
For extended archiving (up to 2 months), remove the slide from the cassette and submerge it in 30% glycerol cryoprotectant at -20 °C .
For Xenium workflow compatibility, once tissue sections are present on a Xenium slide, place slides in a pre-cooled, sealed slide mailer on dry ice (maximum two slides per mailer, ensuring they do not contact each other), transport on dry ice, and store at -80 °C for up to 4 weeks prior to Analyzer processing.
Storage beyond 1 week is best achieved using cryoprotectant as described above to preserve signal integrity and spatial fidelity.
Image Registration
HuBMAP spatial analyses frequently require precise co-registration of highly multiplexed fluorescence data with conventional brightfield H&E pathology images. Because hematoxylin renders nuclei dark in brightfield whereas DAPI labels nuclei with high-intensity fluorescence, the two signals are inversely correlated. The protocol must therefore specify inversion of the DAPI channel prior to registration (e.g., using Fiji/ImageJ “Edit → Invert”) to harmonize contrast polarity and improve feature correspondence.
All registration steps must be exhaustively documented for reproducibility, including software versioning and plugin configuration. For global alignment, the protocol should explicitly define use of the Fiji plugin “Linear Stack Alignment with SIFT (Scale Invariant Feature Transform)” configured to perform an Affine transformation. Affine models provide sufficient degrees of freedom to correct for translation, rotation, isotropic scaling, and shear, which are typical inter-modality discrepancies between Xenium fluorescence optics and brightfield slide scanners.
For tissues exhibiting complex, spatially non-linear distortions (e.g., optical barrel distortion, section compression, or mechanical tearing), the protocol must document elastic (non-rigid) registration.
Specifically, it should detail execution of the bUnwarpJ plugin, including manual specification of the Landmark Weight parameter (e.g., 0.9) to prioritize feature correspondence while preserving biologically realistic tissue morphology. All transformation parameters and interpolation methods must be recorded to ensure computational reproducibility and compliance with reporting standards.
HuBMAP metadata architecture and data provenance
The protocol must also account for the generation and documentation of specific digital files mandated by the HuBMAP directory schema. These files are generated by the micro-meta app and are required to describe the optical physics, numerical apertures, laser intensities, and hardware configurations used during the automated imaging phases of the spatial assay, ensuring that the optical data can be universally standardized across different geographic mapping centers.
| A | B | C | |
| extras/ | microscope_hardware.json, microscope_settings.json | Documents optical configurations, hardware specifications, and laser settings via the micro-meta app. 1 | |
| raw/ | transcript_locations.csv, custom_probe_set.csv, gene_panel.csv | Contains unprocessed coordinate data (x, y, z), target gene IDs, and probe sequence definitions. 2 | |
| lab_processed/ | *.ome.tiff, *ome-tiff.channels.csv, tissue-boundary.geojson | Houses loss-less compressed image pyramids, channel metadata, and manually annotated spatial tissue boundaries. 2 |
microscope_hardware.json: Documents the optical physics and hardware configurations used during imaging.
microscope_settings.json: Records laser intensities, numerical apertures, and specific instrument settings.
transcript_locations.csv: Contains unprocessed coordinate data (x, y, z) and target gene IDs.
custom_probe_set.csv: Defines the sequences used in the targeted padlock probe mix.
gene_panel.csv: Lists the full set of genes targeted in the assay.
morphology.ome.tif: The raw, loss-less compressed image pyramid used for morphological reference and segmentation.
tissue-boundary.geojson: Contains the manually or automatically annotated spatial boundaries of the tissue.
Protocol references
CG000578: Xenium In Situ for FFPE - Tissue Preparation Handbook.
CG000579: Xenium In Situ for Fresh Frozen Tissues - Tissue Preparation Guide.
CG000580: Xenium In Situ for Fresh Frozen Tissues - Fixation & Permeabilization.
CG000582: Xenium In Situ Gene Expression - User Guide (covers Hybridization, Ligation, & Amplification).
CG000623: Xenium Cassette Assembly and Chemical Decrosslinking - User Guide.
CG000749: Xenium In Situ Gene Expression with Cell Segmentation Staining - User Guide.
CG000760: Xenium Prime In Situ Gene Expression with optional Cell Segmentation Staining - User Guide.
Biological & Benchmarking References
Janesick, A., Shelansky, R., Gottscho, A.D., Wagner, F., Williams, S.R., Rouault, M., Beliakoff, G., Morrison, C.A., Oliveira, M.F., Sicherman, J.T. and Kohlway, A., 2023. High resolution mapping of the tumor microenvironment using integrated single-cell, spatial and in situ analysis. Nature communications, 14(1), p.8353.
Marco Salas, S., Kuemmerle, L.B., Mattsson-Langseth, C., Tismeyer, S., Avenel, C., Hu, T., Rehman, H., Grillo, M., Czarnewski, P., Helgadottir, S. and Tiklova, K., 2025. Optimizing Xenium In Situ data utility by quality assessment and best-practice analysis workflows. Nature Methods, 22(4), pp.813-823.
Vannan, A., Lyu, R., Williams, A.L., Negretti, N.M., Mee, E.D., Hirsh, J., Hirsh, S., Hadad, N., Nichols, D.S., Calvi, C.L. and Taylor, C.J., 2025. Spatial transcriptomics identifies molecular niche dysregulation associated with distal lung remodeling in pulmonary fibrosis. Nature genetics, 57(3), pp.647-658.
Technical & Protocol-Specific References
Bava, F.A., CHRISTOPHERSON, C., Costa, J., Gohil, S., Galonska, C. and Nagendran, M., 10X Genomics Inc, 2024. Immobilization methods and compositions for in situ detection. U.S. Patent Application 18/503,949.
Ma, X., Chen, P., Wei, J., Zhang, J., Chen, C., Zhao, H., Ferguson, D., McGee, A.W., Dai, Z. and Qiu, S., 2024. Protocol for Xenium spatial transcriptomics studies using fixed frozen mouse brain sections. STAR protocols, 5(4).
Jain, S., Knoten, A., Lagwankar, A.K. and Reinert, S., 2025. Xenium Spatial Transcriptomics Protocol for Human Kidney.