Tumor microenvironment research requires more than identifying which cells are present. CD Genomics integrates single-cell sequencing, spatial transcriptomics, targeted spatial RNA profiling, and spatial proteomics to connect tumor, immune, stromal, and vascular cell states with their tissue locations and protein phenotypes.
The tumor microenvironment is shaped by local cellular composition, tissue architecture, and molecular signaling. A complete study must connect cell identity with location, neighborhood, and phenotype.
Single-cell RNA sequencing resolves tumor, immune, and stromal states, but it cannot show whether a suppressive macrophage population is located inside the tumor core, at the invasive margin, or beside excluded lymphocytes.
Bulk and low-resolution measurements may detect immune activation, extracellular matrix remodeling, or hypoxia without showing whether the signal originates from tumor cells, fibroblasts, macrophages, endothelial cells, or mixed regions.
Ligand–receptor expression generates candidate interactions. Spatial mapping helps determine whether sender and receiver populations occupy compatible regions, prioritizing interactions supported by both expression and tissue organization.
CD Genomics connects cell-state discovery, tissue-level spatial mapping, and protein validation into a unified TME evidence chain. Each layer addresses the limitation of the previous, reducing inference gaps from discovery through validation.
Cell states, spatial localization, and protein phenotypes are integrated into a single framework — each layer addresses the limitation of the previous.
Niche boundaries and tumor–stroma interfaces are defined by tissue coordinates, not computational inference alone.
Discovery, Mechanism, and FFPE Cohort routes are selected based on specimen type, prior knowledge, and project stage.
Every output is supported by complementary technologies, not single-assay inference. RNA findings are confirmed at the protein level where required.
The project is built as a connected evidence chain. Technologies are selected according to the biological question, specimen type, experimental groups, and the level of validation required.
The solution is not a fixed assay package. Each method fills a specific evidence gap, and not every project requires every module.
| Technology Module | Methods | Role in the Project | Representative Outputs |
|---|---|---|---|
| Cell-State Discovery | Single-cell RNA-seq (10x Chromium), single-nucleus RNA-seq (10x Chromium) | Define tumor, immune, stromal, and vascular states. | Cell atlas, marker genes, abundance and functional-state scores. |
| Immune Profiling | Immune repertoire sequencing, surface-protein profiling | Characterize immune clonality and phenotype. | Clonotype tables, paired receptor data, immune-state profiles. |
| Spatial Transcriptome Mapping | Whole-transcriptome spatial transcriptomics (Visium, Stereo-seq), high-resolution spatial profiling | Map cell states and gene programs across intact tissue. | Spatial domains, cell maps, region-specific expression. |
| Targeted Spatial RNA | In situ RNA profiling (Xenium, CosMx SMI), customized transcript panels | Resolve selected genes at higher spatial resolution. | Cell-resolved RNA maps and co-expression patterns. |
| Spatial Protein Validation | mIHC, multiplex IF, GeoMx spatial proteomics | Validate tumor, immune, and stromal phenotypes. | Protein maps, phenotype counts, proximity measurements. |
| Integrated Bioinformatics | Deconvolution, boundary, neighborhood, and communication analysis | Connect modalities into one spatial evidence chain. | Niche maps, interaction networks, prioritized pathways. |
Choose the starting route according to specimen preservation, prior knowledge, project stage, and the evidence needed for the final research conclusion.
Best suited for: exploratory studies, limited prior knowledge, new cell-state discovery, and construction of a reference TME atlas.
Best suited for: candidate targets, experimental treatment groups, immune exclusion, altered tumor–immune interactions, and RNA-to-protein validation.
Best suited for: archived specimens, pathology-annotated cohorts, limited tissue, candidate spatial markers, and defined research-group comparisons.
Each analysis module is linked to a biological question and a defined set of research outputs.
Map immune cell states to intact tissue and identify immune-rich, immune-excluded, myeloid-dominant, lymphoid-rich, and candidate tertiary lymphoid structure-like regions.
Examine molecular and cellular changes across the tumor core, invasive margin, stromal regions, and adjacent tissue using pathology-guided spatial analysis.
Identify recurrent local communities formed by tumor, immune, stromal, and vascular populations instead of relying only on whole-sample cell proportions.
Combine ligand-receptor expression with tissue regions, neighborhood membership, and spatial proximity to prioritize positionally plausible sender-receiver relationships.
Evaluate activation, cytotoxicity, exhaustion, interferon response, antigen presentation, myeloid programs, and selected immune-checkpoint markers across tissue regions.
Compare treatment and control groups, baseline and post-intervention samples, primary and metastatic models, time points, tissue regions, or target perturbations.
Individual methods answer different parts of the TME question. Integration connects cell identity, tissue location, neighborhood organization, and selected protein phenotypes.
| Research Dimension | Evidence from an Individual Method | Evidence Added by Integration |
|---|---|---|
| Cell Identity and Location | Cell states or tissue expression patterns are measured separately, leaving either location or cellular source uncertain. | Cell-state signatures are connected with tumor regions, invasive margins, stromal areas, and immune niches. |
| Niches and Boundaries | One method may capture morphology, expression, or cell composition without the other evidence layers. | Pathology, spatial expression, and cellular composition jointly support niche and boundary characterization. |
| Cell-Cell Communication | Ligand-receptor co-expression identifies candidate interactions without confirming compatible tissue locations. | Spatial proximity and neighborhood context help prioritize positionally plausible interactions; functional validation may still be required. |
| RNA and Protein Evidence | Transcript and protein measurements may be generated or interpreted separately. | Matched or consecutive tissue sections support regional RNA-protein comparison and selected phenotype validation. |
The final technology combination is selected according to the biological question, specimen type, tissue area, required spatial resolution, cohort design, and validation needs.
Exact requirements depend on tissue type, preservation method, selected platform, section size, and study design. Feasibility should be reviewed before project initiation.
| Research Material | Recommended Starting Route | Main Planning Consideration |
|---|---|---|
| Fresh tumor tissue | Single-cell RNA-seq + matched spatial transcriptomics | Coordinate viable-cell preparation with tissue-section collection. |
| Frozen tumor tissue | Single-nucleus RNA-seq + spatial transcriptomics | Suitable when viable cells cannot be recovered. |
| FFPE tissue block | FFPE spatial RNA + spatial protein profiling | Use pathology review and consecutive tissue sections. |
| Small biopsy | Targeted spatial RNA or protein profiling | Prioritize tissue area, marker selection, and section allocation. |
| Matched experimental groups | Coordinated multi-modal profiling | Standardize collection, region selection, and batch structure. |
| Existing single-cell data | New spatial profiling + data integration | Review data quality, annotation, and reference compatibility. |
From hypothesis definition to integrated interpretation, each stage is connected by assay-specific quality control and documented analysis decisions.
Clarify the hypothesis, groups, tissue regions, pathways, and validation needs.
Assess preservation, replication, pathology information, and matched data.
Choose cell-state, spatial RNA, spatial protein, and optional modules.
Perform coordinated assays with quality-control checkpoints.
Map states, niches, boundaries, pathways, and communication networks.
Provide data, figures, methods, and a structured interpretation report.
The final package is matched to the selected route and analysis scope.
Directly aligns scRNA-seq cell-state clusters with Visium or Xenium spatial coordinates, eliminating assumption-based manual annotation.
mIHC, GeoMx DSP, and CODEX cover whole-tissue to single-cell resolution, enabling protein validation at the scale that matches transcriptomic findings.
Probe-based FFPE spatial transcriptomics enables analysis of archived pathology cohorts without requiring fresh tissue collection.
Discovery, Mechanism, and FFPE Cohort routes each have defined technology-to-output mappings, ensuring the project scope matches the research question.
These studies illustrate how cell-state discovery and spatial mapping can reveal local tumor ecosystems. The case-study visuals below are original conceptual summaries rather than reproductions of published figures.
Cellular diversity can be resolved by single-cell sequencing, but local tissue relationships require spatial profiling.
The study combined single-cell RNA sequencing of tumor and matched non-tumor tissues with spatial transcriptomics of matched tissue sections.
Integrated mapping localized immune and non-immune states and evaluated whether selected ligand–receptor candidates were supported by spatial co-localization.
The work demonstrates how cell-state discovery and tissue mapping provide complementary evidence for local immune organization.
Different tumor subclones may occupy distinct regions and associate with different immune and stromal communities.
Visium spatial transcriptomics identified tissue-wide subclones, while higher-resolution spatial RNA imaging examined local cellular composition.
Distinct subclones were associated with different immune, fibroblast, endothelial, and macrophage populations, together with candidate local signaling.
The study shows how tissue-wide discovery and higher-resolution profiling can prioritize subclone-specific niches and communication mechanisms.
No. It is most valuable when the project requires discovery of new cell states or a sample-specific reference. Projects with defined markers, limited tissue, or suitable references may begin with spatial RNA or protein profiling.
Single-cell sequencing provides detailed molecular profiles but generally removes tissue location. Spatial transcriptomics preserves coordinates but may capture multiple cells per feature. Integration connects cell states with intact tissue architecture.
Yes. Probe-based FFPE spatial transcriptomics and multiplex spatial protein assays can profile archived sections. Feasibility depends on preservation, tissue integrity, area, morphology, block age, and section availability.
The boundary can be defined using pathology annotation, histological segmentation, tumor-marker expression, inferred malignant-cell abundance, or a combination of these approaches.
No. It identifies candidate interactions. Spatial proximity improves biological plausibility, but selected interactions may still require protein, perturbation, co-culture, imaging, or other functional validation.
Some measurements can be combined on the same section. In many projects, RNA and protein assays are performed on consecutive sections and aligned using morphology and computational registration.
Yes. Cell states, domains, neighborhoods, immune programs, and communication networks can be compared when biological replication, region selection, and batch design are planned appropriately.
Yes, when the existing dataset has sufficient quality, metadata, and relevant reference populations. Compatibility depends on tissue, biological condition, handling, species, and sequencing strategy.
Share your tissue type, preservation method, experimental groups, existing data, candidate pathways, and desired outputs. Our team can help determine whether your project should begin with cell-state discovery, spatial mapping, protein validation, or an integrated multi-modal route.
Discuss Your TME StudyOnline Inquiry
