Spatial Omics
Bring omics data into the spatial dimension and reconstruct molecular expression and tissue architecture in situ.
BGI Stereo-Seq (FFPE)
Stereo-seq FFPE spatial transcriptomics is BGI Research's dedicated clinical-sample profiling solution built on the upgraded Stereo-seq platform. It targets ultra-high-resolution spatial molecular profiling of formalin-fixed paraffin-embedded (FFPE) samples. Through an innovative de-crosslinking repair system and a total-RNA capture strategy, it overcomes the RNA degradation and molecular crosslinking bottlenecks of FFPE samples, achieving tri-modal resolution of "tissue morphology – full-transcriptome expression – spatial localization" at 500 nm subcellular resolution, while simultaneously capturing host–microbe in situ interaction trajectories. Over 90% of clinical pathology samples are archived as FFPE, which preserve key disease information yet remain an undeciphered "black box" under conventional methods. This breakthrough awakens the molecular value of global FFPE archives and bridges the gap between clinical samples, spatial omics, and clinical translation. It is widely applied in precision oncology, infectious disease mechanisms, rare-disease pathology, and antibody drug development, providing core support for deploying domestic spatial omics in the clinic.
Principle Introduction
Stereo-seq FFPE centers on "de-crosslinking repair – nanoscale capture – total-RNA coverage – spatial localization." It optimizes multiple steps for FFPE characteristics to resolve low-quality samples efficiently. A gentle de-crosslinking workflow precisely breaks RNA–protein crosslinks and repairs fragmentation from long storage, restoring nucleic-acid activity even for samples with severely degraded RNA (DV200 < 30%). A high-density capture chip with 500 nm spacing arrays spatially barcoded probes, each site mapping to a subcellular region for resolution far beyond conventional methods. Unlike traditional poly(T) probes, a "master-key" random-primer system captures all RNA types (mRNA, lncRNA, snoRNA, etc.) without relying on the poly(A) tail, enabling uniform full-length coverage and aberrant alternative-splicing detection. Captured RNAs receive spatial barcodes, and after reverse transcription, amplification, and library construction combined with high-throughput sequencing and image registration, subcellular spatial molecular maps are generated; species-specific alignment simultaneously reveals host and microbial gene activity in situ.
Literature
Cell published "Stereo-seq V2: Spatial mapping of total RNA on FFPE sections with high resolution," which used the Stereo-seq FFPE core technology to achieve the first high-resolution in situ co-analysis of host and microbe in FFPE samples, solving both clinical-sample profiling and interaction-mechanism challenges. The study included FFPE samples (breast cancer, tuberculous lung tissue) stored 2–9 years, some with DV200 < 30%—low-quality samples intractable to conventional methods. Stereo-seq FFPE mapped fine spatial molecular atlases of cancerous, immune-infiltrated, and necrotic regions, distinguished tumor subtypes, and captured spatially resolved lncRNA/snoRNA expression missed by traditional methods, identifying aberrant alternative-splicing isoforms linked to drug resistance. In tuberculous lung tissue, it co-localized Mycobacterium dynamics with host immune-cell response trajectories, revealing for the first time specific B-cell clones aggregating around infection foci that mutate and mature near the lesion—reconstructing the in situ host–pathogen interplay. The study confirms Stereo-seq FFPE's advantages in low-quality samples, total-RNA capture, and host–microbe co-analysis.
Sample Requirements
tissue block ≥0.3cm×0.3cm×0.2cm, intact with no damage or mold, sealed at room temperature; fixation 6–24 h (4% PFA), avoid under-/over-crosslinking; prefer storage ≤9 yr; note tissue type, pathological grade, and clinical info.
5–10 μm continuous sections, ≥5 per sample, on adhesive slides, air-dried and sealed cold; no wrinkles, detachment, or contamination; include 1 H&E slide for pathology orientation; avoid freeze–thaw before deparaffinization; note section orientation and key pathological regions.
prefer intact samples covering tumor core, invasive margin, and adjacent tissue; avoid necrosis >30%; note tumor type, stage, treatment history, follow-up; flag low-quality samples (DV200<30%) for tailored repair.
prefer lesion–normal boundary; note pathogen type (e.g., Mycobacterium, virus); follow biosafety shipping; storage ≤8 yr to preserve pathogen nucleic acid for host–microbe co-analysis.
attach pathology report and preparation records; 2–3 sections acceptable if scarce, noting value and location; samples >5 yr need RNA-integrity pre-test; customized optimization available.
Deliverables
Raw sequencing data
Fastq reads, spatial-barcode allocation statistics, RNA repair/capture efficiency raw data, sequencing QC (GC, duplication), and nanosite signal-intensity analysis.
Data analysis report
subcellular gene-expression spatial maps, full-RNA (coding/non-coding) analysis, alternative-splicing event identification, host–microbe (optional) in situ co-localization maps, and pathological-region molecular association interpretation with visualizations.
FAQ
Two optimizations: (1) gentle de-crosslinking repairs RNA–protein crosslinks and fragmentation, restoring activity; (2) random primers independent of poly(A) capture all RNA fragments, breaking reliance on intact mRNA—effective even at DV200<30%.
No. Random primers unbiasedly capture all RNA; a dedicated bioinformatics pipeline uses species-specific alignment to auto-distinguish host and microbial (bacterial, viral) transcripts and co-localize them—ideal for tracing pathogens in difficult infectious diseases.
