Frontiers in Spatial Multi-Omics Integration: From Data to Mechanism
InquirySpatial multi-omics integration is opening new ways to study how cells work together within their native environments. By combining information from different molecular layers—such as gene expression, protein distribution, and epigenetic modifications—researchers can uncover patterns that were invisible when studying each layer separately. This article highlights recent advances in multi-modal spatial technologies, the main computational challenges in bringing diverse datasets together, and how these approaches are driving new insights into disease mechanisms.
Multi-Modal Technology Landscape
Spatial multi-omics is not just about collecting more data—it's about linking different types of molecular information in the same spatial context. This "technology matrix" allows researchers to match patterns across genomics, transcriptomics, proteomics, and epigenomics, revealing how molecular layers interact in tissues.
What is a spatial multi-modal technology matrix?
A spatial multi-modal technology matrix is a set of methods that capture different molecular features from the same tissue location. For example, it may combine transcriptomic data with protein or epigenetic measurements, enabling direct cross-comparison within the same cell or tissue region.
Epigenetics + Transcriptome: snmC-seq3
snmC-seq3 is designed to profile both DNA methylation and chromatin accessibility alongside transcriptomic information at single-nucleus resolution. In the adult mouse brain, this approach has been used to generate comprehensive maps that connect epigenetic states to gene expression patterns in spatially defined cell populations. Such data help decode how cell identity and function are regulated in specific tissue niches.
The workflow of dissection, nuclei and library preparation for snmC-seq3 and snm3C-seq. (Liu, H.,et al., Nature, 2023).
Protein + Transcriptome Co-Detection: EpicIF
EpicIF enables researchers to measure protein expression while also capturing RNA transcripts from the same section. This synchronous detection provides a more complete molecular profile of the tissue microenvironment. For example, correlating immune marker proteins with the corresponding gene expression patterns can clarify immune cell states in tumor or inflammatory regions.
How can proteins and transcripts be measured at the same time?
Methods like EpicIF achieve this by using spatial barcodes or imaging-based detection for proteins, followed by sequencing-based readouts for RNA from the same sample, ensuring the two datasets are directly aligned in space.
Emerging Platforms and Technology Trends
Spatial multi-omics continues to evolve rapidly, with new platforms pushing the boundaries of resolution, throughput, and the number of molecular layers measured simultaneously. These innovations aim to make spatial profiling more comprehensive, more scalable, and applicable to a wider range of research samples.
High-Coverage Spatial Transcriptomics
Recent advances in techniques like Slide-seqV2 and Stereo-seq have significantly improved spatial resolution and transcript capture efficiency. These platforms allow researchers to profile tens of thousands of transcripts across whole tissue sections, enabling fine-grained mapping of rare cell populations or subtle microenvironmental changes.
A summary of the study. The spatial transcriptomics (Stereo-seq, 53 samples) and RNA-seq (16 samples) acquisition workflow for 23 patients with liver cancer were analyzed as the discovery cohort. (Wu, L., et al., Cell, 2023).
Multi-Modal Imaging-Sequencing Hybrids
New hybrid platforms integrate high-plex imaging with sequencing readouts. For example, imaging mass cytometry can visualize dozens of protein markers at subcellular resolution, while a paired spatial transcriptomics workflow captures RNA from the same region. This combination yields a deep molecular portrait, linking protein localization with transcriptional activity in situ.
Low-Input and Archival Sample Compatibility
Emerging protocols are being optimized for formalin-fixed paraffin-embedded (FFPE) tissues and other low-quality or scarce samples. This is particularly valuable for studies using clinical archives, rare animal models, or limited biobank material, as it expands the scope of spatial multi-omics without requiring fresh tissue.
Triple-Layer and Beyond: RNA + Protein + Epigenetics
While most current methods combine two modalities, early demonstrations of RNA + protein + chromatin accessibility in a single assay are showing promise. These workflows aim for "all-in-one" spatial molecular profiling, where researchers can simultaneously track gene regulation, transcriptional output, and protein abundance in the same cellular context.
The field is moving toward higher resolution, larger spatial coverage, increased molecular layers per assay, and compatibility with diverse sample types, all supported by better computational tools for integration and visualization.
Computational Integration Challenges
While new spatial multi-omics methods are producing increasingly rich datasets, combining these diverse data types into a single, interpretable framework remains a major hurdle. Differences in measurement techniques, data formats, and spatial resolutions create integration barriers that require advanced computational solutions.
Key challenges include:
- Batch effects — technical variations between experiments that can mask true biological signals.
- Modality mismatch — datasets collected with different resolutions or detection methods may not align perfectly in space or scale.
- Data sparsity — some technologies produce incomplete or low-coverage data, making integration less straightforward.
Correcting Batch Effects: Seurat v5
Seurat v5 is a widely used toolkit for integrating multi-omics data. Its weighted nearest neighbor (WNN) approach helps align datasets from different batches or modalities by identifying common cell or tissue states. For spatial applications, this means RNA and epigenetic maps—or RNA and protein maps—can be more reliably compared.
Following quality control analysis of RNA-Seq data, cells were clustered using Seurat's default integration workflow, identifying eight distinct clusters. (Abdallah, Ali T., et al., International Journal of Molecular Sciences, 2024)
Mapping Non-Spatial Data into Spatial Contexts: SIMO
SIMO (Spatial Integration of Multi-Omics) uses probabilistic alignment to project non-spatial single-omics datasets—such as ATAC-seq or DNA methylation—onto spatial reference maps. This allows researchers to infer where specific molecular states are located within a tissue, even if the original data lacked spatial coordinates.
Modality-Independent Embeddings: SpaMosaic
SpaMosaic applies graph neural networks and contrastive learning to create modality-agnostic spatial embeddings. This not only corrects batch effects but also enables "missing modality imputation," where unmeasured molecular layers can be predicted from available data.
Overview of SpaMosaic framework. (Chen, Jinmiao, et al., 2024)
How do researchers correct batch effects in multi-omics data?
They use statistical and machine learning methods, such as WNN integration, mutual nearest neighbor matching, or graph-based embeddings, to remove technical variation while preserving biological differences.
Breakthrough Case: Alzheimer's Disease and the Amyloid Plaque Microenvironment
In Alzheimer's disease (AD), amyloid-β (Aβ) plaques are distinctive pathological features that assemble into dense aggregates. Recent spatial multi omics studies have enabled unprecedented insights into the molecular makeup surrounding these plaques, revealing coordinated changes in proteins, lipids, and transcriptomic signatures within tissue microenvironments.
Integrated Proteomics and Lipid Imaging in Human AD Brains
Toyama and colleagues applied matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) combined with laser microdissection (LMD)-based proteomics and lipid imaging on post-mortem brain sections of AD patients. This integrated spatial multi-omics approach uncovered dynamic distributions of Aβ proteoforms, key cytoskeletal proteins (MAP1A, MAP1B, MAP2), and dysregulated lipid species in plaque-associated regions. Lipid–protein co-localization highlighted the significant role of lipid metabolism in modulating plaque development and neurotoxicity.
High-Resolution Spatial Transcriptomics around the Plaque Niche
Another study deployed two advanced spatial transcriptomics platforms—CosMx and Stereo seq—to map gene expression changes and cellular interactions in the Aβ plaque microenvironment in AD mouse models. The analysis revealed altered cellular compositions, including microglia-astrocyte signaling disruptions, transcriptional activation of immune and inflammatory pathways, and early glial responses that potentially contribute to plaque maintenance or clearance.
Resolving the plaque niche using spatial transcriptomics. (Mallach, Anna, et al. Cell Reports, 2024).
Broader Review of Spatial Multi-Omics Applications in AD
A comprehensive review of spatial multi-omics applications in AD highlighted how combining transcriptomics, proteomics, metabolomics, and epigenomics has deepened our understanding of spatial molecular heterogeneity. The review emphasizes that integrating diverse modalities provides clearer insights into how AD-associated pathology, including plaques and tangles, reorganizes cellular neighborhoods and affects regional molecular landscapes.
Key Takeaways
| Aspect | Insight |
|---|---|
| Spatial Integration Value | Multi-modal mapped data around plaques reveal interactions between proteins, lipids, and cellular responses in situ. |
| Techniques in Action | MALDI-MSI + proteomics/lipid imaging provide real-world molecular state maps; CosMx & Stereo-seq enable high-resolution cell-type and transcriptome-specific localization. |
| Mechanistic Clarity | Spatial patterns of lipid dysregulation and glial activation suggest plaque-associated microenvironments are active sites of pathogenesis, not inert deposits. |
How Spatial Multi-Omics Helps Illuminate Alzheimer's Disease Pathology
Spatial multi-omics integrates layers like transcriptomics, proteomics, metabolomics, and epigenomics with precise spatial context. In Alzheimer's disease (AD), this integration clarifies how molecular alterations and cellular interactions occur around hallmark lesions—such as amyloid plaques and neurofibrillary tangles—offering insights that traditional bulk or dissociated single-cell methods cannot reveal.
Mapping Molecular and Cellular Neighborhoods
Spatial methods allow visualization of where cell types and molecular signals organize around pathological features. For instance, coordinating multiple omics modalities across AD tissue highlights how immune, neuronal, and glial elements cluster and respond near amyloid plaques.
Revealing Heterogeneity within the Lesion Microenvironment
Spatial omics identifies distinct cellular and molecular states—such as reactive glial subtypes or dysregulated metabolic pathways—that concentrate near plaques or vascular amyloid depositions. Such heterogeneity is key to deciphering localized disease dynamics.
Bridging Molecular Layers with Spatial Precision
By combining imaging with molecular profiling—such as spatial transcriptomics with mass spec proteomics and metabolomics—researchers can directly link gene expression changes to protein modifications or lipid variations in situ. This provides a multi-layered perspective on plaque-associated pathology.
Advancing Resolution, Throughput, and Tissue Coverage
Modern spatial multi-omics platforms achieve whole-transcriptome profiling or high-plex protein imaging across wide tissue areas, capturing thousands of molecular features without sacrificing spatial resolution. These improvements enable comprehensive mapping of AD-affected brain regions while preserving fine-grained context.
Summary Table
| Insight Area | Contribution of Spatial Multi-Omics in AD |
|---|---|
| Spatial cellular context | Integrates where specific cell types and molecules reside relative to lesions |
| Microenvironment heterogeneity | Differentiates molecular states around plaques, aiding mechanistic understanding |
| Multi-layer linkage | Correlates transcript, protein, lipid, and epigenetic data in a spatial context |
| Enhanced tech capability | Enables high-resolution, high-throughput mapping across large AD brain areas |
Partner with Us for Your Spatial Multi-Omics Research
At CD Genomics, we specialize in delivering high-quality spatial multi-omics solutions designed to meet the needs of research institutions and laboratories worldwide. Our expertise spans multi-modal assay design, advanced data integration, and custom bioinformatics pipelines—helping you move from complex datasets to meaningful biological insights.
Whether you are mapping cellular microenvironments in neurodegenerative diseases, exploring cross-modal molecular interactions, or developing new hypotheses about disease mechanisms, our team provides the technical depth and collaborative approach you need. With rigorous quality control and end-to-end project support, we ensure your data are accurate, reproducible, and ready for publication.
Contact us today to discuss your project requirements and discover how our spatial multi-omics capabilities can accelerate your next breakthrough.
Reference
- Liu, H., Zeng, Q., Zhou, J. et al. Single-cell DNA methylome and 3D multi-omic atlas of the adult mouse brain. Nature 624, 366–377 (2023).
- Wu, L., Yan, J., Bai, Y. et al. An invasive zone in human liver cancer identified by Stereo-seq promotes hepatocyte–tumor cell crosstalk, local immunosuppression and tumor progression. Cell Res 33, 585–603 (2023).
- Abdallah, Ali T., and Anna Konermann. "Unraveling divergent transcriptomic profiles: A comparative single-cell RNA sequencing study of epithelium, gingiva, and periodontal ligament tissues." International Journal of Molecular Sciences 25.11 (2024): 5617.
- Yang, Penghui, et al. "Spatial integration of multi-omics single-cell data with SIMO." Nature communications 16.1 (2025): 1265.
- Chen, Jinmiao, et al. "Mosaic integration of spatial multi-omics with SpaMosaic." (2024).
- Mallach, Anna, et al. "Microglia-astrocyte crosstalk in the amyloid plaque niche of an Alzheimer's disease mouse model, as revealed by spatial transcriptomics." Cell Reports 43.6 (2024).
- Ma, Yixiao, et al. "Spatial multi-omics in alzheimer's disease: a multi-dimensional approach to understanding pathology and progression." Current Issues in Molecular Biology 46.5 (2024): 4968-4990.
- Marshall, Cody R., et al. "Untangling Alzheimer's disease with spatial multi-omics: a brief review." Frontiers in Aging Neuroscience 15 (2023): 1150512.
- Toyama, Yumiko, et al. "Integrated spatial multi-omics study of postmortem brains of Alzheimer's disease." Acta Histochemica et Cytochemica 57.3 (2024): 119-130.
