Spatial Omics

Mapping disease cells in tissue is essential, but extracting and analysing them is what advances drug discovery. Spatial cell sorting enables physical isolation of individual cells for comprehensive multi-omics profiling – bridging the gap between tissue visualisation and functional validation

Hee-Sool Rho at Meteor Biotech

Spatial omics: from maps to mechanisms

Over the past decade, spatial omics technologies have revolutionised biological research, evolving from creating high-resolution tissue maps to uncovering mechanistic insights that drive precision medicine forward. While early platforms excelled at generating detailed atlases showing where cells reside and how they organise into functional niches, they often fell short of explaining the ‘why’ behind location-dependent cellular behaviours. Today, the field is experiencing a fundamental shift – moving beyond static description towards dynamic understanding by directly interrogating cells within their native microenvironments.1

This evolution reflects a paradigm shift in biological discovery. Rather than homogenising tissue samples and losing critical spatial context, researchers recognise the need to integrate spatial information with deep molecular profiling. The global spatial omics market, valued at $403m in 2024, is projected to reach $974m by 2033, underscoring the field’s rapid expansion and clinical relevance.2 This growth is driven by increasing recognition that spatial context holds the key to understanding complex biological processes, from tumour microenvironment dynamics to immune cell infiltration patterns.3 Within this landscape, spatial cell sorting platforms are emerging as essential tools for converting spatial omics from passive observation into an active engine for mechanistic discovery.

The critical gap: from visualisation to actionability

Despite remarkable advances in spatial transcriptomics and imaging, the field faces a fundamental challenge: converting visually rich data into functionally testable hypotheses. Current commercial spatial profiling platforms excel at mapping gene expression patterns and detecting phenomena such as metabolic gradients or hypoxic zones at tumour margins.1,4 However, these technologies remain primarily observational, unable to physically isolate specific cells or microregions for comprehensive downstream analysis.4 Array-based systems provide valuable spatial information but face resolution limitations – each capture spot typically contains multiple cells, diluting single-cell precision.4 Meanwhile, multiplexed in situ imaging platforms deliver exquisite subcellular resolution and can visualise hundreds to thousands of transcripts simultaneously.1 These platforms are advantageous for analysing large tissue sections, yet the economic burden increases substantially when processing many samples, limiting their scalability for high-throughput studies. Furthermore, these technologies fundamentally rely on pre-designed probe panels targeting known genes, meaning they can only confirm previously characterised transcripts. Novel gene expression, unexpected pathway activation and unannotated transcripts remain invisible. These platforms also cannot resolve critical molecular details such as adenosine-to-inosine (A-to-I) RNA editing events, alternative splicing patterns, or full-length isoform diversity – features increasingly recognised as key determinants of drug resistance and therapeutic response.1,5

This creates a paradox; we can see disease-driving cells with unprecedented clarity, but cannot extract the deep molecular information needed for drug response testing, target validation or comprehensive biomarker discovery. What the field truly needs is the ability to perform full-length transcriptome analysis on spatially defined cell populations – a capability that would unlock an entirely new dimension of biological insight. There is a growing need for new technologies that can address these limitations, combining spatial precision with comprehensive molecular profiling at economically viable scales.

As noted in recent research, this actionability gap represents one of the most important limitations preventing spatial biology from reaching its full potential in drug development.6 Translational programmes in oncology and immunology increasingly encounter bottlenecks where critical cellular subpopulations can be visualised but not molecularly characterised at the depth required for clinical decision-making. Addressing these challenges requires fundamentally new approaches to spatial analysis.

Spatial cell sorting: bridging the actionability gap

Spatial cell sorting represents a fundamentally new technology category that combines high-resolution spatial mapping with physical cell isolation. These systems enable researchers to physically separate individual cells from tissue while preserving spatial identity and molecular integrity – a capability that has eluded the field for decades.

The spatial cell sorting workflow integrates three critical steps. First, visualisation using standard or advanced staining methods (haematoxylin & eosin, immunofluorescence, or multiplex immunohistochemical) identifies regions of interest based on morphology, biomarkers or artificial intelligence (AI)-driven segmentation. Second, precision extraction mechanisms gently isolate selected cells into collection wells while preserving molecular integrity. Various technological approaches, such as near-infrared laser-based systems, have been developed to minimise sample damage compared to conventional UV-based technologies, ensuring superior RNA integrity essential for downstream full-length sequencing.4

Crucially, because spatial cell sorting physically retrieves a single cell or intact cells, rather than capturing fragmented molecules in situ, it enables true full-length transcriptome analysis. Researchers can now detect alternative splicing events, characterise complete isoform repertoires, identify fusion transcripts and profile RNA modifications – all from cells whose precise tissue location is known. When combined with parallel genomic and proteomic workflows, spatial cell sorting creates the foundation for integrated multi-omics analysis at the level of individual cells or defined microregions.

Beyond RNA analysis, spatial cell sorting platforms demonstrate exceptional versatility in DNA-based genomic profiling. Recent advances have integrated spatial cell sorting with barcoded multiple displacement amplification, enabling high-coverage whole-genome sequencing that captures copy number alterations, single nucleotide variants, structural variations and kataegis signatures from spatially defined tissue microniches.7 Workflows combining spatial cell sorting-based isolation with cytopathological profiling have successfully reconstructed clonal evolution in complex blood cancers, demonstrating the ability to analyse archival bone marrow samples and resolve genomic heterogeneity across haematopoietic lineages.8

Research applications and translational potential

The translational potential of spatial cell sorting is becoming increasingly evident through published research demonstrating its utility in biomarker discovery and tumour microenvironment analysis. A growing body of work illustrates how spatially informed isolation, when coupled with full-length transcriptomics, can reshape the questions addressed in disease research.

The research applications of spatial cell sorting were demonstrated in published studies and subsequently highlighted in a review article, where researchers developed the novel sequencing methodology; an innovative method combining spatial cell sorting with full-length transcriptomics to profile cancer stem cell microniches in residual tumours from triple-negative breast cancer patients following neoadjuvant chemotherapy.5,9 This approach enabled simultaneous analysis of gene expression counts, alternative splicing variations, B-cell and T-cell receptor sequences, and characterisation of the A-to-I editome – epitranscriptomic features impossible to detect with conventional spatial techniques that rely on short RNA fragments.

Remarkably, the Select-seq study identified that the frequency of A-to-I editing in the glutathione peroxidase 4 (GPX4) gene – characteristic of cancer stem cells exhibiting highly immunosuppressive gene expression – could stratify patients into two distinct groups. Those with cancer stem cells showing higher degrees of A-to-Iediting in GPX4 had significantly lower chances of survival, demonstrating spatial cell sorting’s potential for discovering clinically actionable biomarkers.5 Such epitranscriptomic biomarkers would remain entirely hidden without the combination of spatial precision and full-length transcript analysis. AI-driven image analysis increasingly segments tissues into detailed microenvironments, enabling researchers to build detailed mechanism-of-action maps while accelerating target selection.10

Toward integrated multi-omics platforms for precision diagnostics

The true power of spatial cell sorting emerges when genomic, transcriptomic and proteomic data are integrated from the same spatially defined cell populations. Each molecular layer provides complementary information: DNA sequencing reveals mutations, copy number changes and structural variants that drive disease; full-length RNA analysis uncovers how these genetic alterations manifest as aberrant splicing, novel isoforms or RNA modifications; and proteomic profiling confirms which molecular changes translate into functional protein expression. When all three layers are captured from cells whose tissue context is precisely known, a comprehensive picture of disease biology emerges that no single-omics approach can provide.

This integrated approach has profound implications for diagnostics and therapeutic decision-making. Consider a tumour where imaging identifies a therapy-resistant niche at the invasive margin. Spatial cell sorting could isolate cells specifically from this region, enabling parallel analysis of their mutational profile, splicing patterns and protein markers. If these cells harbour a specific splice variant associated with drug resistance, or express a particular protein isoform that could be targeted therapeutically, this information would directly inform treatment strategy – yet it would be invisible to conventional spatial or bulk sequencing methods.

AI-driven pathology requires high-quality spatial labels where imaging features and molecular profiles are precisely matched at single-cell resolution. Spatial cell sorting approaches generate ideal ground-truth data sets for training predictive models.11 Industry-wide, there is growing interest in building spatial multi-omics reference cohorts that can underpin regulatory-grade AI tools for precision diagnostics.

Future horizons: organoids, multiplexed therapeutics and clinical translation

The spatial biology market’s next expansion phase will be driven by three converging trends: advanced 3D model systems; multiplexed therapeutic approaches; and direct clinical implementation. Patient-derived organoids and co-culture systems are becoming central platforms for disease modelling and drug screening, yet their inherent heterogeneity demands spatial resolution to capture critical cell-cell interactions and niche-specific responses.3 Spatial cell sorting enables spatially resolved sampling within these complex 3D structures, allowing separate analysis of invasion fronts, hypoxic cores or immune-infiltrated regions. This capability supports high-throughput organoid drug screening with spatial readouts and temporal mapping of how combination therapies reshape cellular ecosystems. Multiplexed therapeutic strategies demand spatial resolution to understand how different components of a regimen reshape specific cell populations over time. Spatially guided isolation can compare molecular states before and after treatment in situ, guiding the design of combinations that selectively dismantle pathogenic niches.

Looking towards clinical translation, spatial cell sorting could integrate into routine pathology workflows. Future scenarios might involve pathologists identifying high-risk regions in tumour resections, applying spatial cell sorting for targeted cell enrichment and performing rapid multi-omics profiling to inform treatment decisions. When full-length transcriptome, genome and proteome data from spatially defined cells become available within clinically relevant time frames, precision oncology will move from population-level statistics to truly individualised therapy selection based on the specific molecular architecture of each patient’s disease.

Conclusion

The transition from spatial mapping to spatial action marks a defining moment in precision medicine. The combination of spatial cell sorting with full-length transcriptomics and integrated multi-omics analysis opens new frontiers in biomarker discovery – enabling detection of splicing variants, RNA modifications and isoform diversity that conventional approaches cannot capture. With the ability to integrate genomic, transcriptomic and proteomic information from precisely defined tissue locations, the field stands ready to unlock the next era of precision medicine.

As spatial cell sorting technologies mature and integrate with AI-driven analysis, the actionability gap will increasingly close, enabling mechanism-based drug discovery and personalised therapeutic strategies.

References:

1. Moffitt J R et al (2022), ‘The emerging landscape of spatial profiling technologies’, Nat Rev Genet, 23(12), 741-759
2. https://www.marketdataforecast.com/market-reports/spatial-omics-market
3. Bressan D et al (2023), ‘The dawn of spatial omics’, Science, 381, eabq4964
4. Lee S et al (2024), ‘Mapping cancer biology in space: applications and perspectives on spatial omics for oncology’, Mol Cancer, 23(26)
5. Lee A C et al (2022), ‘Spatial epitranscriptomics reveals A-to-I editome specific to cancer stem cell microniches’, Nat Commun, 13(2540)
6. Liu L et al (2024), ‘Spatiotemporal omics for biology and medicine’, Cell, 187(17), 4488-4519
7. Kim J et al (2023), ‘Barcoded multiple displacement amplification for high coverage sequencing in spatial genomics’, Nat Commun, 14(5261)
8. Jeong D et al (2023), ‘Hema-seq reveals genomic aberrations in a rare simultaneous occurrence of hematological malignancies’, Cell Rep Methods, 3(100617)
9. Lee AC (2023), ‘Epitranscriptomics of cancer microniches’, Nat Rev Cancer, 23(139)
10. Lee Y et al (2022), ‘Derivation of prognostic contextual histopathological features from whole-slide images of tumours via graph deep learning’, Nat Biomed Eng, 6(12), 1452-1466
11. Schroeder A et al (2025), ‘Scaling up spatial transcriptomics for large-sized tissues: uncovering cellular-level tissue architecture beyond conventional platforms with iSCALE’, Nat Methods, 22(9), 1911-1922