26 Single-Cell and Spatial Omics Atlases Worth Bookmarking

Inquiry

A single-cell or spatial omics atlas is a curated reference map that organizes cell identities, molecular states, and, when available, tissue locations across defined biological systems. Newly generated datasets often benefit from comparison with biologically appropriate reference resources because atlases can support cell-type identification, label transfer, cross-cohort comparison, external benchmarking, and hypothesis generation. Their value is not that they replace a project-specific dataset, but that they provide a structured context for asking whether an observed cell state, marker, or spatial pattern is expected, unusual, or reproducible.

However, atlas resources differ in species, tissue sampling, anatomical depth, disease state, donor composition, assay modality, tissue handling, and annotation strategy. Choosing a useful reference therefore requires biological congruence first and technical compatibility second. For repository discovery and download workflows, see our companion guide on Databases and Resources: How to Find Spatial Omics Datasets and Protocols. For decisions among bulk, cell-resolved, and spatial profiling, see Bulk RNA-seq vs Single-Cell RNA-seq vs Spatial Transcriptomics: How to Choose for Tissue Studies.

This guide compiles 26 single-cell and spatial atlas resources across major organ systems and explains when each type of reference is useful. The list includes both maintained atlas portals and study-specific atlas datasets; not every item is a standalone database product.

Key Takeaways:

• Curated Atlas vs Data Repository: An atlas adds organized curation, harmonized annotation, and a reusable biological reference framework, whereas a repository primarily archives study-level raw and/or processed datasets and metadata.
• Targeted Selection Hierarchy: Prioritize biological compatibility (species, anatomical subregion, age, and disease state) before matching single-cell or spatial technology platforms.
• Healthy Baseline vs Disease Atlases: Healthy tissue atlases establish physiological homeostatic references, whereas disease atlases are essential for capturing transitional, activated, or exhausted cell states.
• Computational Utility: Reference atlases empower automated label transfer, marker prioritization, and spatial transcriptomics deconvolution, but they do not automatically remove cohort-specific batch artifacts.
• Research-Use Alignment: External reference integration provides robust exploratory hypotheses and discovery pipelines for preclinical and basic life-science investigations.

What Is a Single-Cell or Spatial Omics Atlas?

In single-cell genomics, a reference atlas is a systematically organized representation of cellular diversity within a biological system. Some atlases are broad healthy-tissue references; others focus on disease, development, one anatomical region, one lineage, or cross-species comparison. Atlas scale also varies: a deeply profiled study may include a limited donor set, while integrated community resources can span hundreds of individuals or many independent studies. The defining feature is not size alone, but a reusable annotation and comparison framework.

Understanding the operational difference between a data repository and a curated atlas is important for study design. Repositories such as GEO, SRA, EGA, GSA-Human, or ArrayExpress can store raw reads, processed matrices, metadata, and study-level annotations. A reference atlas goes further by organizing datasets into a biologically interpretable map with curated cell identities, harmonized labels, or spatial reference coordinates. An atlas may still rely on a repository for its underlying files, and researchers should retain the original study metadata when reusing either resource.

How to Choose the Right Reference Atlas for Your Study

Selecting a poorly matched reference can distort downstream interpretation. For example, a healthy adult atlas may correctly identify broad stromal or immune lineages in a pediatric tumor sample yet fail to represent malignant, developmental, treatment-induced, or transitional states. Use a reference as a comparison framework rather than a source of forced labels, and review mismatches before deciding whether label transfer, integration, or de novo annotation is appropriate.

1. Biological and Anatomical Congruence: Match species, organ, anatomical subregion, developmental stage, and sex when these variables are relevant to the biology.
2. Condition and Perturbation Alignment: Decide whether the project needs a healthy baseline, a disease-matched reference, a treatment-exposed cohort, or a longitudinal atlas.
3. Assay Modality and Tissue Handling: scRNA-seq, snRNA-seq, chromatin, imaging, and spatial assays capture different molecular compartments and can introduce different sampling biases. Nuclei-based references may better represent some frozen or difficult tissues but do not reproduce whole-cell RNA composition.
4. Cell Metadata and Label Granularity: Check whether the atlas provides the resolution required for the study, from broad lineages to disease-associated states, vascular subtypes, immune states, or anatomical niches.
5. Computational Accessibility: Confirm whether raw counts, processed matrices, metadata, coordinates, and reusable analysis objects such as .rds or .h5ad files are available and whether access is open or controlled.
6. Intended Analysis: Choose an atlas that supports the actual downstream task: annotation, external validation, cross-study integration, spatial mapping, model comparison, or hypothesis generation.

Quick Selection Matrix

Access status is part of atlas selection, not an administrative detail. Public portals may provide processed matrices and metadata while raw human sequencing files remain controlled because of consent or privacy requirements. A project should document which layer is actually available: web visualization, processed expression matrices, reusable analysis objects, raw reads, or cohort-level metadata. If only a paper or accession record is accessible, treat that resource as a study reference rather than assuming it can support full reanalysis. For cross-study integration, the most useful atlas is usually the one with enough metadata to reconstruct donor, tissue, condition, and technical covariates—not simply the resource with the largest cell count.

Figure 2. Strategic decision flowchart: evaluating biological congruence, single-cell modality, and computational metadata before integrating reference atlas datasets.

Whole-Body Reference Atlases

Organism-wide reference atlases establish broad cellular baselines across multiple organ systems, making them ideal for initial cell lineage identification, pan-tissue comparisons, and cross-organ biomarker screening.

1. Human Cell Atlas (HCA) Data Portal

• Scope: International initiative building reference maps of human cells across tissues, life stages, and contributing projects.
• Modality: Project-dependent single-cell, single-nucleus, spatial, and multi-omic datasets.
• Best Used For: Broad human reference mapping, cell-identity benchmarking, and discovery of tissue-specific expression patterns.
• Access / Primary Resource: Human Cell Atlas Data Explorer. Access conditions vary by contributing project.

2. Single Cell Atlas

• Scope: Multi-omics reference portal focused on healthy human fetal and adult tissues.
• Modality: Single-cell transcriptomic, spatial, chromatin-accessibility, and other atlas/query modules available through the portal.
• Best Used For: Exploring healthy tissue cell types, gene expression, cell-type distributions, and cross-modality reference patterns.
• Access / Primary Resource: Single Cell Atlas (public web portal).

3. Tabula Sapiens

• Scope: Nearly 500,000 cells from 24 human tissues and organs, with multiple tissues collected from the same donors. [1]
• Modality: Droplet-based and plate-based single-cell RNA sequencing.
• Best Used For: Cross-tissue comparison within shared donors, broad healthy-cell reference mapping, and evaluation of tissue-specific transcriptional programs.
• Access / Primary Resource: Tabula Sapiens Data Portal (public portal).

Brain and Nervous System Atlases

The central and peripheral nervous systems exhibit extreme cellular diversity, complex alternative splicing, and strict spatial compartmentalization, making specialized brain atlases essential for neurobiological research.

4. BRAIN Initiative Cell Atlas / BICCN

• Scope: Molecular and anatomical brain-cell references spanning human, non-human primate, and mouse resources; the adult human transcriptomic atlas provides a major cell-type reference. [3]
• Modality: Single-cell/single-nucleus transcriptomics, epigenomics, and spatial mapping across BRAIN Initiative projects.
• Best Used For: Deep neuronal and glial classification, brain-region comparison, regulatory annotation, and cross-species neurobiology.
• Access / Primary Resource: BICCN Data Portal (public project portal).

5. Multiregion Single-Cell Atlas of Alzheimer's Disease

• Scope: Disease-focused human brain resource integrating multiple regions and neuropathologically characterized Alzheimer's disease cohorts.
• Modality: Primarily single-nucleus transcriptomics with cohort-specific supporting data.
• Best Used For: Comparing region-specific neuronal vulnerability, glial activation, and disease-associated cell states across neurodegenerative brain regions.
• Access / Primary Resource: AD Knowledge Portal. Access varies by contributing cohort and dataset.

6. Glioblastoma Spatial Transcriptomics Atlas / Study Resource

• Scope: Spatially resolved human glioma dataset spanning IDH-wild-type glioblastoma and IDH-mutant glioma samples, including spatially annotated tumor regions.
• Modality: 10x Visium spatial transcriptomics with integrative spatial analysis.
• Best Used For: Studying spatially organized glioma cell states, hypoxic core-to-edge organization, and local tumor-state neighborhoods.
• Access / Primary Resource: GEO GSE237183 (public series with processed data).

Respiratory System Atlases

Respiratory reference atlases support the dissection of mucosal immunity, structural remodeling, viral infection dynamics, and chronic inflammatory pulmonary diseases.

7. Human Lung Cell Atlas (HLCA)

• Scope: Integrated human lung reference combining more than 2.4 million cells across healthy and diseased samples from multiple studies. [2]
• Modality: Integrated scRNA-seq and snRNA-seq reference data.
• Best Used For: Reference mapping, label transfer, rare lung-cell identification, and comparison of healthy versus disease-associated lung states.
• Access / Primary Resource: CZ CELLxGENE HLCA Collection (public collection).

8. COVID-19 Immune Atlas

• Scope: Large-scale single-cell immune landscape of COVID-19 across peripheral blood and respiratory samples from patients and controls.
• Modality: Single-cell RNA sequencing with cohort-level immune-state analysis.
• Best Used For: Comparing systemic and respiratory immune states, myeloid responses, lymphocyte programs, and severity-associated immune changes.
• Access / Primary Resource: GSA-Human HRA001149 (controlled access).

9. Bronchoalveolar Immune Atlas

• Scope: Bronchoalveolar lavage fluid atlas from healthy controls and patients with moderate or severe COVID-19.
• Modality: scRNA-seq with paired TCR-seq in selected samples.
• Best Used For: Investigating local lung immune composition, macrophage states, tissue-resident T cells, and severity-associated bronchoalveolar responses.
• Access / Primary Resource: GEO GSE145926 (public).

10. Lethal Lung Atlas

• Scope: Autopsy-derived lung atlas comparing fatal COVID-19 lungs with control lungs using rapidly collected frozen tissue.
• Modality: Single-nucleus RNA sequencing of lung tissue.
• Best Used For: Studying end-stage lung injury, epithelial and stromal remodeling, immune infiltration, and cell states associated with lethal disease.
• Access / Primary Resource: GEO GSE171524 (public processed data; linked study resources describe raw-data access).

11. Post-Tuberculosis Lung Atlas

• Scope: Human post-tuberculosis lung study resolving persistent senescence, inflammatory programs, and vascular/endothelial changes after microbiological clearance. [11]
• Modality: scRNA-seq with matched bulk RNA-seq in the study resource.
• Best Used For: Investigating long-term cellular remodeling after tuberculosis and separating persistent post-infectious states from acute infection signatures.
• Access / Primary Resource: GSA-Human HRA004142 and HRA004156 (controlled access).

Cardiovascular System Atlases

Cardiovascular research demands robust single-nucleus methods due to the extreme physical dimensions and fragility of adult cardiomyocytes and the dense extracellular matrix of vascular walls.

12. Dilated and Hypertrophic Cardiomyopathy Atlas

• Scope: Human heart single-nucleus resource comparing non-failing hearts with dilated and hypertrophic cardiomyopathy and linking pathogenic variants to cell composition and transcriptional programs. [6]
• Modality: snRNA-seq with genetic and transcriptomic integration.
• Best Used For: Comparing cardiomyocyte stress programs, fibroblast activation, vascular states, and cell-type-specific effects of cardiomyopathy-associated variants.
• Access / Primary Resource: Primary Science publication; data access is described in the paper.

13. Atherosclerotic Plaque Spatial Atlas

• Scope: Spatially resolved single-cell transcriptome atlas of human atherosclerotic plaques, including plaque tertiary lymphoid organ-like structures. [10]
• Modality: scRNA-seq and Stereo-seq spatial transcriptomics.
• Best Used For: Resolving vascular and immune cell niches, smooth-muscle-cell states, B-cell-rich tertiary lymphoid structures, and spatial plaque organization.
• Access / Primary Resource: GSA-Human HRA006030 (open access).

Digestive System Atlases

Gastrointestinal and hepatic tissues exhibit steep metabolic and microbial gradients, requiring reference atlases that capture epithelial crypt-villus axes, zonation, and tumor-stroma boundaries.

14. Multimodal Spatial Cell Atlas of Intestine

• Scope: Multimodal human intestinal reference mapping cell types and tissue organization across the intestine. [5]
• Modality: Single-cell/single-nucleus transcriptomics combined with spatial and multiplexed imaging modalities.
• Best Used For: Mapping epithelial differentiation, immune/stromal neighborhoods, crypt-villus organization, and tissue-compartment relationships in healthy intestine.
• Access / Primary Resource: HuBMAP Data Portal (public portal; dataset-level access varies).

15. Pancreatic Ductal Adenocarcinoma (PDAC) Atlas

• Scope: Single-cell transcriptomic resource from primary PDAC and normal pancreas samples in the foundational study dataset.
• Modality: scRNA-seq.
• Best Used For: Examining malignant ductal heterogeneity, stromal and immune composition, fibroblast states, and tumor-versus-normal cell programs in PDAC.
• Access / Primary Resource: GSA CRA001160 (study data resource).

16. Spatial Atlas of Liver Cancer

• Scope: Human liver-cancer resource profiling tumor regions and an invasive zone at spatial resolution.
• Modality: Stereo-seq spatial transcriptomics with matched single-cell transcriptomic analysis in the study.
• Best Used For: Studying tumor-hepatocyte crosstalk, local immunosuppression, invasive-zone biology, and spatial heterogeneity in liver cancer.
• Access / Primary Resource: CNGB CNSA CNP0002199 (study data resource).

17. Evolutionary Atlas of Esophageal Squamous Cell Carcinoma (ESCC)

• Scope: Study resource integrating precancerous lesions, primary ESCC, and spatial tumor contexts to investigate disease evolution.
• Modality: scRNA-seq and spatial transcriptomics.
• Best Used For: Tracing epithelial-state transitions, stromal and immune remodeling, and spatial programs across precancer-to-cancer progression.
• Access / Primary Resource: SpatialMapESCC GitHub resource (public analysis resource).

Figure 3. Organ-system architecture: major anatomical reference resources spanning neurological, cardiovascular, pulmonary, digestive, and renal systems.

Kidney and Urinary System Atlases

The kidney comprises highly specialized nephron segments with dramatic functional and physiological divergence, necessitating spatially anchored reference atlases.

18. Multimodal Spatial Atlas of Human Kidney

• Scope: Human kidney atlas resolving healthy and injured cell states and tissue niches across nephron and stromal compartments. [4]
• Modality: Single-nucleus transcriptomics with spatial and complementary molecular profiling in the KPMP/HuBMAP ecosystem.
• Best Used For: Kidney cell-type annotation, injury-state mapping, nephron-segment comparison, and spatial localization of disease-associated cell states.
• Access / Primary Resource: KPMP Kidney Tissue Atlas Explorer (public explorer; some underlying datasets may have additional access conditions).

19. Cross-Species Single-Cell Kidney Atlas

• Scope: Integrated atlas of more than one million kidney cells from human, mouse, and rat samples across healthy and disease contexts. [9]
• Modality: Integrated snRNA-seq across human and rodent kidneys with cross-species harmonization.
• Best Used For: Comparing conserved cell states and evaluating how closely rodent kidney models reproduce human cell-type and pathway-level disease programs.
• Access / Primary Resource: SISKA data repository on Zenodo (public data resource).

Immune and Hematologic Atlases

Because immune cells circulate and adapt to distinct tissue microenvironments, pan-cancer and lineage-specific immune atlases are indispensable for translational immunology and oncology research.

20. Pediatric AML Longitudinal Atlas

• Scope: Longitudinal pediatric AML atlas spanning diagnosis, remission, and relapse from patients representing multiple disease subtypes.
• Modality: Single-cell RNA-seq and single-cell ATAC-seq.
• Best Used For: Studying treatment-associated cellular plasticity, relapse-state transitions, and changes in malignant and normal hematopoietic hierarchies.
• Access / Primary Resource: EGA study EGAS00001007323 (controlled access).

21. Pan-Cancer T-Cell Atlas (TCellMap)

• Scope: Reference map of 308,048 T cells from 486 samples and 324 individuals across 16 cancer types, with detailed T-cell state annotation. [12]
• Modality: Integrated scRNA-seq with spatial validation and an interactive T-cell reference map.
• Best Used For: Comparing tumor-infiltrating T-cell states, including exhausted, regulatory, proliferative, follicular-helper, and stress-response programs across cancers.
• Access / Primary Resource: TCellMap portal (public processed-data and visualization resource).

22. Pan-Cancer Natural Killer (NK) Cell Atlas

• Scope: Pan-cancer single-cell panorama of NK cells integrating tumor and normal-tissue NK-cell states across multiple cancer types. [8]
• Modality: Integrated scRNA-seq; study data include tumor-infiltrating NK-cell profiles.
• Best Used For: Investigating NK-cell heterogeneity, tumor-associated dysfunction, tissue-resident programs, and conserved cytotoxic states across cancers.
• Access / Primary Resource: GEO GSE212890 (public processed data; raw data are linked from the record).

23. Pan-Cancer B-Cell Atlas

• Scope: Pan-cancer single-cell resource characterizing B-cell and plasma-cell states across multiple tumor types.
• Modality: scRNA-seq with B-cell-state and tumor-microenvironment analysis.
• Best Used For: Comparing naive, memory, germinal-center-like, plasma-cell, and tertiary-lymphoid-structure-associated B-cell programs across cancers.
• Access / Primary Resource: GEO GSE233236 (public study resource).

24. Curated Cancer Cell Atlas (3CA)

• Scope: Curated and continuously expanded cancer scRNA-seq resource spanning many tumor types, studies, samples, and malignant/non-malignant cell states. [7]
• Modality: Curated, standardized scRNA-seq datasets with analysis and visualization resources.
• Best Used For: Cross-study cancer-cell comparison, recurrent transcriptional program analysis, context-dependent gene expression, and intratumoral heterogeneity research.
• Access / Primary Resource: Curated Cancer Cell Atlas (public exploration and downloads; individual source-dataset permissions may vary).

Developmental and Connective-Tissue Atlases

Transient embryonic niches and ubiquitous stromal elements require cross-tissue reference frameworks to disentangle baseline function from pathological activation.

25. SARS-CoV-2 Placental Niches Atlas

• Scope: Human placental study integrating maternal-fetal tissue architecture with infection-associated spatial niches.
• Modality: Visium spatial transcriptomics integrated with published single-cell and single-nucleus placenta references.
• Best Used For: Investigating maternal-fetal tissue compartments, infection-associated niches, trophoblast environments, and local immune organization in placenta.
• Access / Primary Resource: GEO GSE222987 (public).

26. Cross-Tissue Fibroblast Atlas

• Scope: Integrated fibroblast atlas spanning synovium, intestine, lung, and salivary gland across inflammatory disease and control samples.
• Modality: Integrated scRNA-seq with cross-tissue reference mapping.
• Best Used For: Distinguishing shared inflammatory fibroblast states from tissue-specific programs and comparing fibroblast activation across chronic inflammatory diseases.
• Access / Primary Resource: Primary open-access publication and data links.

What Can You Actually Do With a Reference Atlas?

Public reference atlases provide actionable utility across multiple stages of an omics research workflow:

• Cell-Type Annotation & Label Transfer: Tools such as Seurat Reference Mapping, Symphony, SingleR, and related methods can project a query dataset onto an annotated reference. The transferred labels should still be reviewed against markers, dataset quality, and biological context.
• Spatial Transcriptomics Deconvolution and Mapping: Matched single-cell references can support deconvolution or cell mapping when spatial measurements do not directly resolve every cell identity. The required method depends on platform resolution, segmentation, and tissue context. See How to Integrate scRNA-Seq with Spatial Transcriptomics for Cell-Type Deconvolution and Spatial Mapping.
• Cross-Cohort Harmonization & Replication: An atlas can help test whether a candidate cell state or marker recurs in independent donors, studies, tissues, or disease cohorts. Integration should preserve study-level metadata and account for technical and biological differences.
• Marker Prioritization & In Silico Pre-Screening: Atlas portals can be used to assess whether a candidate receptor, pathway, or marker is enriched in the intended cell population before new experiments are designed. RNA-level evidence does not establish protein abundance, target engagement, or functional effect.
• Cross-Species Model Evaluation: Cross-species atlases can identify conserved and divergent cell states before a rodent model is used to support a human research question. Similarity in transcriptomic state does not by itself establish equivalent drug response or disease biology.

To implement rigorous computational workflows with external reference data, explore our specialized Single-Cell Omics Bioinformatics and Data Mining Service and dedicated Cell Type Mapping Services.

When an Atlas Is Not Enough

While reference atlases provide immense value, researchers must recognize their scientific and technical boundaries:

1. Demographic and Anatomical Mismatches: A reference atlas derived from healthy adult organ donors cannot fully capture pediatric developmental states, extreme elderly frailty, or specific anatomical micro-niches.
2. Absence of Disease-Specific Novel States: Standard healthy atlases do not contain neoplastic subclones, therapy-resistant plastic phenotypes, or severe pathogen-induced metaplasias. Attempting to force query cells into an incomplete reference can obscure novel biology.
3. Batch and Platform Confounders: Differences in dissociation chemistry, sequencing chemistry (e.g., 10x 3' v3 vs 5' v2 vs Smart-seq2), and sequencing depth frequently introduce technical batch effects that require careful algorithmic harmonization rather than direct merging.
4. Lack of Spatial Coordinates in Dissociated Atlases: Most public single-cell resources profile dissociated single-cell or single-nucleus suspensions, discarding spatial coordinates and extracellular architectural context.

Figure 4. Computational integration pipeline: utilizing reference atlases for automated label transfer, spatial deconvolution, cross-cohort replication, and experimental follow-up.

From Atlas Lookup to Public-Data Mining

An atlas lookup becomes a true public-data mining project when the goal shifts from finding a reference to testing a mechanistic question across datasets. A defensible workflow starts by defining the claim and evidence needed, then curates biologically comparable cohorts, preserves donor- and study-level metadata, applies quality review and batch-aware integration, and uses only the downstream analyses required to test that claim. Candidate findings should then be challenged in an independent cohort or by orthogonal experimental evidence rather than treated as validated because they recur in an integrated embedding.

When a project requires systematic dataset discovery, cross-study harmonization, deeper cell-state analysis, regulatory or interaction modeling, and a validation plan, see our Single-Cell Omics Bioinformatics and Data Mining Service. For routine processing of a newly generated dataset, the Single-Cell RNA-Seq Data Analysis Service remains the more direct workflow.

FAQs

Advancing Your Single-Cell and Spatial Research Projects

Selecting an appropriate reference atlas can strengthen cell-type annotation, cross-cohort comparison, spatial mapping, and hypothesis generation. When public atlases do not capture the tissue context, perturbation, disease state, species, or molecular modality required by a study, new single-cell or spatial data may be needed to answer the question directly.

CD Genomics supports research-use single-cell and spatial omics projects from study-design review through sequencing and bioinformatics integration. Explore our Single-Cell Sequencing Services and Single-Cell RNA-Seq Data Analysis Service when new data generation or project-specific analysis is required. Service scope and sample feasibility are confirmed for each research project before execution.

References

1. The Tabula Sapiens Consortium. The Tabula Sapiens: A multiple-organ, single-cell transcriptomic atlas of humans. Science. 2022;376(6594):eabl4896.
2. Sikkema L, Ramírez-Suástegui C, Strobl DC, et al. An integrated cell atlas of the lung in health and disease. Nature Medicine. 2023;29(6):1563–1577.
3. Siletti K, Hodge R, Mossi Albiach A, et al. Transcriptomic diversity of cell types across the adult human brain. Science. 2023;382(6667):eadd7046.
4. Lake BB, Menon R, Winfree S, et al. An atlas of healthy and injured cell states and niches in the human kidney. Nature. 2023;619(7970):585–594.
5. Hickey JW, Becker WR, Nevins SA, et al. Organization of the human intestine at single-cell resolution. Nature. 2023;619(7970):572–584.
6. Reichart D, Lindberg EL, Maatz H, et al. Pathogenic variants damage cell composition and single cell transcription in cardiomyopathies. Science. 2022;377(6606):eabo1984.
7. Tyler M, Gavish A, Barbolin C, et al. The Curated Cancer Cell Atlas provides a comprehensive characterization of tumors at single-cell resolution. Nature Cancer. 2025;6(6):1088–1101.
8. Tang F, Li J, Qi L, et al. A pan-cancer single-cell panorama of human natural killer cells. Cell. 2023;186(19):4235–4251.e20.
9. Klötzer KA, Abedini A, Li S, et al. Analysis of individual patient pathway coordination in a cross-species single-cell kidney atlas. Nature Genetics. 2025;57:1922–1934.
10. Lai Z, Kong D, Li Q, et al. Single-cell spatial transcriptomics of tertiary lymphoid organ-like structures in human atherosclerotic plaques. Nature Cardiovascular Research. 2025;4:547–566.
11. Sun G, Li K, Ping J, et al. A single-cell transcriptomic atlas reveals senescence and inflammation in the post-tuberculosis human lung. Nature Microbiology. 2025;10(8):2073–2091.
12. Chu Y, Dai E, Li Y, et al. Pan-cancer T cell atlas links a cellular stress response state to immunotherapy resistance. Nature Medicine. 2023;29:1550–1562.

Research Use and Trust Statement

This resource is intended for research planning and educational use. Atlas content, access conditions, dataset composition, and portal availability can change over time, so researchers should verify the current project page, accession record, and primary publication before reuse. CD Genomics services are for research use only and are not intended for clinical diagnosis, patient management, or therapeutic decision-making.