Spatial Transcriptomics Sample Preparation Guide: Tissue Handling, Sectioning, and Quality Control Requirements

Inquiry

Figure 1: The two major sample preparation paths for spatial transcriptomics — fresh frozen and FFPE — with key decision points at tissue collection, sectioning, and QC.

Sample preparation is the single largest source of variability in spatial transcriptomics — and the most difficult to reverse. A poorly frozen block, an over-fixed FFPE specimen, or a skipped permeabilization optimization can turn a well-funded project into an uninterpretable dataset. This guide covers the practical decisions researchers face before a spatial transcriptomics library is ever made: fresh frozen versus FFPE, tissue collection and embedding, cryosectioning and microtomy, RNA quality metrics and what they actually predict, section quality assessment, permeabilization optimization, and pre-submission QC. The goal is to help research teams move from "we have tissue" to "we have tissue ready for spatial transcriptomics" with fewer failed runs and less wasted budget.

Fresh Frozen vs. FFPE Paths

The first decision in spatial transcriptomics sample preparation is also the most consequential: fresh frozen or FFPE. The choice determines capture chemistry, QC metrics, sectioning equipment, and whether permeabilization optimization is required.

Fresh frozen tissue generally yields higher RNA integrity and is compatible with a wider range of platforms, including Stereo-seq and Visium fresh frozen. FFPE tissue is far more abundant — biobanks and pathology archives are dominated by FFPE blocks — and is supported by Visium FFPE, Visium HD with CytAssist, Xenium, CosMx SMI, and Stereo-seq OMNI. The fresh frozen vs. FFPE decision guide covers the trade-offs in greater depth.

The practical rule: if you have both options, choose fresh frozen for discovery-phase projects where maximum transcript sensitivity matters. Choose FFPE when working with archived cohorts, when histological detail is paramount, or when the platform you have selected is FFPE-optimized.

Fresh Frozen Tissue Step by Step

Fresh frozen tissue preparation follows a defined sequence. Each step has failure modes that are well-characterized and largely avoidable.

Collection and freezing. Speed is the dominant variable. Tissue should be collected and frozen within 30 minutes of excision. Rinse briefly with ice-cold PBS or sterile saline to remove blood, blot dry with lint-free wipes, and trim to fit the capture area — typically 6.5 x 6.5 mm for Visium, up to 13.2 x 13.2 cm for Stereo-seq.

The recommended freezing method is a liquid-nitrogen-cooled isopentane bath. Immerse the tissue for 1–2 minutes until fully frozen. Do not plunge tissue directly into liquid nitrogen — the Leidenfrost effect creates an insulating gas pocket, producing uneven freezing and ice crystal artifacts that destroy morphology.

OCT embedding. Pre-chill OCT compound on wet ice to eliminate bubbles. Fill a cryomold with a thin layer of OCT, position the tissue with the plane of interest facing the bottom, and cover completely. Remove any bubbles near the tissue surface — they create voids in sections. Freeze the block on a pre-chilled metal block on dry ice or in the isopentane bath. Store at -80 C.

Cryosectioning. Equilibrate the OCT block to cryostat temperature (-18 C to -22 C for standard tissues; -25 C to -30 C for fatty tissues) for at least 30 minutes. Cut at 10 um thickness. Discard the first 2–3 sections to clear surface artifacts. Mount sections onto the capture surface by brief thaw-warming. Fix immediately in -20 C methanol for 30 minutes.

Permeabilization optimization. This is the step most frequently skipped — and the most common cause of failed fresh frozen runs. Different tissues release mRNA at different rates, and there is no universal permeabilization time.

The optimization protocol uses a test chip with 4 consecutive sections from the same block, tested at different time points. Published starting points include 12 minutes for mouse brain, 18 minutes for mouse embryo (E12.5–E14.5), 3 minutes for zebrafish embryo, and 24 minutes for human liver and kidney. These are starting points only — tissue age, fixation quality, and cellular density all shift the optimum. The correct time is the one that produces the strongest fluorescent signal with the sharpest tissue footprint, indicating efficient mRNA release without lateral diffusion.

For researchers working with fresh frozen tissue, CD Genomics spatial transcriptomics services cover the full workflow from tissue QC through library preparation and sequencing. Platform-specific guidance is available for Stereo-seq and 10x Genomics Visium fresh frozen workflows.

FFPE Tissue Handling and Sectioning

FFPE sample preparation has fewer variables than fresh frozen — permeabilization is pre-optimized, and sectioning uses standard microtomy — but the variables that do exist are less forgiving.

Fixation. The single most important variable in FFPE spatial transcriptomics is fixation quality. Under-fixation leaves residual enzymatic activity that degrades RNA during storage. Over-fixation creates excessive crosslinks that block probe access and reduce transcript detection. The consensus recommendation is 10% neutral buffered formalin for less than 24 hours total fixation time. Blocks fixed for weeks or months in formalin — common in some clinical pathology workflows — consistently underperform in spatial transcriptomics.

Sectioning. Cut at 5 um on a microtome with a fresh blade. Float sections on a water bath at approximately 38 C. For challenging tissues — dense fibrotic specimens, calcified tissues, fatty samples — SCHOTT NEXTERION Slide H hydrogel-coated slides improve section adherence and prevent detachment during decrosslinking.

Drying and decrosslinking. A graduated drying schedule minimizes cracking and detachment: 42 C for 3 hours, 37 C for 16 hours (overnight), then 60 C for 1 hour. Deparaffinize in xylene or Sub-X (2 x 10 minutes), rehydrate through graded ethanol, then decrosslink at 85 C for 30 minutes in Tris-based buffer. This decrosslinking step is what makes FFPE RNA accessible to capture probes — incomplete decrosslinking is a common cause of low transcript recovery in FFPE spatial experiments.

Special tissue considerations. For calcified or mineralized tissues, traditional EDTA decalcification takes days to weeks and severely compromises RNA integrity. A 2025 protocol using Morse's solution (22.5% formic acid with 10% sodium citrate) achieved decalcification in under 24 hours while preserving transcript recovery at over 5,000 genes per spot — compared to approximately 200 genes per spot with EDTA. For fatty tissues, lower the water bath temperature slightly and consider 4 um sections for cleaner morphology.

For projects starting with FFPE specimens, CD Genomics FFPE spatial transcriptomics provides sample QC, library preparation, and sequencing optimized for formalin-fixed archival tissue.

Figure 2: RNA quality metrics for spatial transcriptomics — RIN (fresh frozen) and DV200 (FFPE) with threshold guidance and the practical reality that even borderline samples can yield meaningful data.

RNA Quality Metrics That Matter

RIN and DV200 are the two numbers that determine whether a spatial transcriptomics project proceeds — but they predict less than many researchers assume.

RIN (RNA Integrity Number) is measured by capillary electrophoresis on RNA extracted from a test cryosection. A value of 7 or above is optimal for fresh frozen workflows. Values between 4 and 7 are marginal — transcript detection sensitivity decreases, and spatial coverage may become uneven, but rejection of all samples below 7 is overly conservative. Values below 4 carry high risk of failure.

The most common causes of RIN degradation are slow tissue collection, delayed freezing, repeated freeze-thaw cycles of the OCT block, RNase contamination during sectioning, and prolonged storage at -80 C.

DV200 is the percentage of RNA fragments longer than 200 nucleotides, measured on RNA extracted from FFPE scrolls. A value of 30% or above is the standard threshold for most FFPE spatial workflows. The Stereo-seq V2 random-primer workflow has produced interpretable data from blocks with DV200 as low as 18%.

A 2026 review of over 1,000 spatial samples emphasized that even below-threshold RIN and DV200 values can yield biologically meaningful data — but the risk of reduced sensitivity and patchy spatial coverage increases. The practical implication is that borderline samples should not be discarded automatically, but they should trigger a pilot section to verify that the workflow produces acceptable data before committing the full sample set.

Beyond the numbers. RIN and DV200 are extracted-RNA metrics — they measure what happens in a homogenized lysate, not at a specific coordinate on the capture surface. Two samples with identical RIN can produce different spatial data quality. DAPI or ssDNA nuclear staining provides complementary information: punctate, well-defined nuclear staining across the section is a strong positive indicator, while diffuse, washed-out staining suggests degradation that RIN alone may miss.

A 2025 Cell Systems commentary emphasized that variation in sample quality "dramatically shapes biological resolution" — high-quality samples resolve subtle cell subtypes and activation states, while low-quality samples may only distinguish coarse lineages such as lymphoid from myeloid. H&E morphological assessment should always accompany RNA QC: check for necrosis covering more than 30% of the region of interest, freezing artifacts, folding, tearing, and detachment.

Section Quality and Common Defects

Section quality assessment precedes RNA QC in the spatial transcriptomics workflow. A section that looks wrong under the microscope will almost certainly produce poor data, regardless of RIN or DV200.

What a good section looks like. The section should be flat, intact, and free of wrinkles, folds, tears, and chatter artifacts — periodic thickness variations that appear as alternating light and dark bands. Adhesion to the capture surface or slide should be complete, with no lifting at edges. No air bubbles or OCT residue should be visible on the tissue surface. For platforms with a defined capture area, the tissue must fall entirely within the active region with no overhang.

Common defects and their causes. Wrinkling and folding typically result from a dull blade, incorrect cryostat temperature, or static electricity. Chatter artifacts indicate that the blade is loose, the block is too cold, or the sectioning speed is too fast. Tissue detachment during processing — particularly during decrosslinking in FFPE workflows — is more common with dense, collagen-rich, or calcified tissues; hydrogel-coated slides reduce this risk.

Tissue-specific challenges. Brain tissue is lipid-rich and structurally heterogeneous; lateral diffusion of mRNA during permeabilization is a particular concern in hippocampus and olfactory bulb. Lung and other air-filled organs collapse during embedding unless perfused with OCT-PBS mixture before freezing. Fibrotic and collagen-rich tissues — cirrhotic liver, cardiac scar, desmoplastic tumors — section unevenly; a fresh blade, slower sectioning speed, and slightly thicker sections (12 um fresh frozen) reduce chatter. Plant tissues present cell walls, starch granules, wax, and air chambers that make embedding and sectioning extremely challenging; additional enzyme treatment during tissue removal may be necessary.

For guidance on handling multiple small specimens on a single capture area, see the tissue multiplexing guide for spatial transcriptomics.

Shipping, Storage, and Pre-Submission QC

Spatial transcriptomics projects routinely involve shipping tissue from a collection site to a sequencing facility. The logistics are unglamorous but consequential — a block that arrives thawed has lost its RNA integrity regardless of how carefully it was prepared.

Storage conditions. OCT-embedded fresh frozen blocks should be stored at -80 C, wrapped in aluminum foil and sealed in airtight bags. Avoid repeated freeze-thaw cycles — each cycle degrades RIN. FFPE blocks should be stored at 4 C in a dry environment. Cut FFPE sections on slides should be stored with desiccant and used within one week when possible.

Shipping. Fresh frozen blocks must ship on sufficient dry ice — at minimum 5 kg for overnight delivery — with blocks wrapped in parafilm to prevent direct ice contact. FFPE blocks ship at 4 C with cold packs. FFPE sections ship at room temperature in sealed slide mailers with desiccant.

Pre-submission QC checklist. Before shipping tissue for spatial transcriptomics, complete these checks:

• Fresh frozen: RIN measured and documented (target >= 7)
• FFPE: DV200 measured and documented (target >= 30%)
• Adjacent H&E or DAPI section confirms intact morphology
• No necrosis covering more than 30% of the region of interest
• No freezing artifacts in fresh frozen sections
• Section thickness confirmed (10 um fresh frozen, 5 um FFPE)
• Tissue dimensions fit within the platform capture area
• Block and section IDs documented with photographs
• For fresh frozen: permeabilization optimization completed and optimal time recorded
• Shipping manifest includes all sample identifiers, tissue type, preservation date, and QC values

A 2025 Nature Biotechnology study from the Spatial Touchstone consortium demonstrated that harmonized sample handling protocols significantly reduce cross-institutional technical variability in spatial transcriptomics — reinforcing that standardized preparation, not just platform choice, determines data quality.

For researchers planning a spatial transcriptomics project from start to finish, the spatial transcriptomics project design guide covers biological question definition, platform selection, and replicate strategy — the logical precursor to the sample preparation steps described here. For Stereo-seq-specific protocols, the Stereo-seq sample preparation and QC guide provides platform-tailored details.

Figure 3: Pre-submission QC checklist — the essential checks to complete before shipping tissue for spatial transcriptomics library preparation.

Sample quality is one of the most important factors affecting spatial transcriptomics success. If you are unsure whether your tissue type, preservation method, or QC metrics meet spatial profiling requirements, CD Genomics experts can help evaluate your samples and recommend an appropriate preparation strategy before project initiation.

FAQ

Q: Can I use samples with RIN below 7 or DV200 below 30% for spatial transcriptomics?

A: Yes, but with acknowledged risk. Even below-threshold samples can yield biologically meaningful data, particularly with newer random-primer workflows that are more tolerant of RNA fragmentation. The trade-off is reduced transcript detection sensitivity and potentially uneven spatial coverage. The practical recommendation is to run a pilot section through the full workflow before committing the entire sample set — a borderline QC value is not an automatic rejection, but it warrants verification that the output meets your project's analytical requirements.

Q: How do I optimize permeabilization for my specific tissue type?

A: Permeabilization optimization requires a test chip with 4 consecutive sections from the same OCT block, tested at different enzyme incubation times. Published starting points exist for common tissues — 12 minutes for mouse brain, 18 minutes for mouse embryo, 3 minutes for zebrafish embryo — but these are calibration points, not guarantees. The optimal time is the one that produces the strongest fluorescent signal with the sharpest tissue footprint. If your tissue type has no published starting point, test a broad range (6, 12, 18, and 30 minutes) and select the condition with the best signal-to-footprint ratio.

Q: How quickly does tissue need to be frozen after collection for fresh frozen spatial transcriptomics?

A: Within 30 minutes is the standard target. RNA degradation begins immediately after tissue devascularization, and the rate accelerates at room temperature. Tissues collected during surgery or biopsy should be transferred to ice-cold PBS or saline immediately after excision, trimmed, and frozen as quickly as feasible. If a delay beyond 30 minutes is unavoidable, document it — RIN measured from the actual block is more informative than elapsed time alone.

Q: What are the most common mistakes in spatial transcriptomics sample preparation?

A: The top four are: skipping permeabilization optimization for fresh frozen workflows, using direct liquid nitrogen immersion instead of an isopentane bath for freezing, over-fixing FFPE tissue (more than 24 hours in formalin), and failing to run a pilot section before committing the full sample set. Each of these is avoidable with standard protocols, and each has caused preventable run failures in published and unpublished spatial transcriptomics projects.

Q: How should I ship tissue samples to a spatial transcriptomics service provider?

A: Fresh frozen OCT blocks must ship on sufficient dry ice — at minimum 5 kg for overnight delivery — with blocks individually wrapped in parafilm and sealed in airtight bags. FFPE blocks ship at 4 C with cold packs. FFPE sections on slides ship at room temperature in sealed slide mailers with desiccant. Every shipment should include a manifest listing block or section IDs, tissue type, preservation date, and RIN or DV200 values. Confirm the shipping protocol with your service provider before sending — platform-specific requirements may apply.

References

1. Grases D, Porta-Pardo E. A practical guide to spatial transcriptomics: lessons from over 1000 samples. Trends in Biotechnology. 2026;44(5):1230-1242. doi:10.1016/j.tibtech.2025.08.020
2. Moffitt JR, Li M, Nie Q, et al. What is the main bottleneck in deriving biological understanding from spatial transcriptomic profiling?. Cell Systems. 2025;16(2):101200. doi:10.1016/j.cels.2025.101200
3. Plummer JT, Dezem FS, Cook DP, et al. Standardized metrics for assessment and reproducibility of imaging-based spatial transcriptomics datasets. Nature Biotechnology. 2025. doi:10.1038/s41587-025-02811-9
4. Jiang Y, Li Y, Cheng M, et al. Protocol for acquiring high-quality fresh mouse lung spatial transcriptomics data. STAR Protocols. 2024;5(1):102825. doi:10.1016/j.xpro.2023.102825
5. Miao J, Ma M, Guo Y, Qin L, Yao L. A practical protocol of processing mineralized tissue for Visium spatial transcriptomics. Mechanobiology in Medicine. 2025;3(4):100163. doi:10.1016/j.mbm.2025.100163

The information provided in this article is for research use only and is not intended for use in diagnostic or therapeutic procedures. CD Genomics provides sequencing and bioinformatics services for research purposes. Researchers should consult the appropriate regulatory guidelines for their specific applications.