Single-Cell Atlases of Ageing Tissues

Single-cell atlases are reference maps built from many individual cells or nuclei. In ageing research, they are used to ask whether an old tissue differs because its cell mixture has changed, because the same cell types have adopted new molecular states, or because both processes are occurring together. The Tabula Muris Senis project demonstrated this logic across mouse tissues by mapping cell-specific ageing changes and age-related shifts in organ cellular composition. [1]

Who This Is Useful For

This page is useful for readers who encounter single-cell ageing studies and want to understand what an atlas can and cannot show. It is especially relevant when interpreting claims about vulnerable cell types, inflammatory cell states, senescence-like signatures, or tissue-specific ageing patterns. [1] [2] [3]

What an Atlas Adds

Bulk tissue measurements remain informative, but they average signals across many cells. A single-cell atlas can distinguish a stronger gene-expression program within a cell type from an increased abundance of that cell type. In the ageing lung, for example, integration of single-cell transcriptomics with proteomics identified cell-type-specific ageing effects, altered epithelial proportions, transcriptional noise, and extracellular-matrix remodeling. [2]

Multi-tissue atlases extend this idea by comparing many organs under a common framework. Tabula Muris Senis profiled 23 mouse tissues and organs across the lifespan, while the Aging Fly Cell Atlas mapped ageing in whole Drosophila at single-nucleus resolution. Both studies support the view that ageing is expressed differently across cell types and tissues. [1] [3]

Composition Versus Cell State

A central distinction in single-cell ageing work is composition versus state. Composition refers to the relative abundance of cell populations, such as immune, epithelial, stromal, endothelial, or progenitor cells. Cell state refers to molecular programs within a population, such as stress-response, inflammatory, metabolic, or senescence-associated expression. Ageing atlases often report both kinds of change. [1] [2] [7]

What Single-Cell Atlases Commonly Measure

Human and Model-Organism Atlases

Model-organism atlases are valuable because tissues can be sampled under controlled conditions and at defined ages. Mouse and fly atlases have therefore been important for studying cell-type-resolved ageing across many tissues. These systems do not make human ageing interchangeable with mouse or fly ageing, but they provide structured maps for comparing conserved and species-specific patterns. [1] [3]

Human ageing atlases are more directly relevant to human biology but face stronger constraints from donor variability, disease history, tissue access, medication exposure, postmortem interval, and uneven sampling across ages. The Human Cell Aging Transcriptome Atlas aggregated human single-cell datasets from 76 publications, covering more than 50 tissue types and ages from 0 to 103 years, which illustrates both the scale and heterogeneity of the available human evidence. [4]

Why Methods Matter

Single-cell datasets are shaped by experimental and computational choices. Quality control, normalization, feature selection, dimensionality reduction, clustering, annotation, and differential expression analysis can all affect the resulting map. Best-practice reviews emphasize that single-cell interpretation depends on the full analysis workflow rather than a single visualization. [5]

Atlas-scale integration adds another layer of interpretation. Batch correction is needed when datasets span protocols, laboratories, donors, or sequencing runs, but benchmarking studies show a tradeoff: integration methods can reduce technical variation while also risking loss of meaningful biological variation. This is especially relevant when ageing, tissue, donor, and disease signals are partly confounded. [6]

Evidence Quality and Interpretation

Confidence is strong that single-cell atlases have improved the ability to identify cell-type-specific ageing signatures and distinguish cell-composition changes from cell-state changes. This conclusion is supported by multi-tissue mouse data, organ-specific studies such as the ageing lung atlas, and organism-wide work in flies. [1] [2] [3]

Confidence is lower when an atlas is used by itself to infer causality. Many ageing atlases are cross-sectional, and technical factors such as dissociation sensitivity, nuclei-versus-cell protocols, sparse gene detection, donor heterogeneity, and batch effects can influence the apparent ageing signal. Methodological papers therefore support interpreting atlas findings alongside spatial data, perturbation experiments, proteomics, histology, and functional assays where available. [5] [6] [9]

What This Does Not Mean

• It does not mean bulk tissue data are obsolete; bulk measurements can still capture system-level signals that are relevant to physiology. [2]
• It does not mean every age-associated cluster is causal; some differences may be secondary, compensatory, or technical. [5]
• It does not mean one atlas provides a universal map of ageing; tissue source, protocol, species, and donor context shape the result. [4] [6]
• It does not mean cell identity is fixed; ageing can blur, shift, or diversify cell states within broad cell-type categories. [7]

Practical Interpretation Examples

• If an old tissue shows more immune transcripts: The signal may reflect more immune cells, stronger immune programs within resident cells, or both. [1] [2]
• If one cell type has a strong ageing signature: That identifies a responsive or vulnerable compartment, not proof that this cell type drives the whole tissue phenotype. [5]
• If two atlases disagree: Differences in species, age range, sampling, dissociation, sequencing protocol, and integration method may all contribute. [6] [9]

Summary

Single-cell atlases have shifted ageing research from tissue averages toward cell-type-resolved maps of composition, state, and molecular programs. Their main value is interpretive: they help clarify which cells contribute to an ageing signal and whether that signal is local, cell-specific, or shared across tissues. Their limits are equally important, because atlas findings remain sensitive to sampling, technical processing, computational integration, and validation outside the sequencing dataset. [1] [5] [6]

References

1. Tabula Muris Consortium. "A single-cell transcriptomic atlas characterizes ageing tissues in the mouse." Nature (2020). https://www.nature.com/articles/s41586-020-2496-1
2. Angelidis, I. et al. "An atlas of the aging lung mapped by single cell transcriptomics and deep tissue proteomics." Nature Communications (2019). https://www.nature.com/articles/s41467-019-08831-9
3. Lu, T.-C. et al. "Aging Fly Cell Atlas identifies exhaustive aging features at cellular resolution." Science (2023). https://pubmed.ncbi.nlm.nih.gov/37319212/
4. Bartz, J. et al. "Human Cell Aging Transcriptome Atlas (HCATA): a single-cell atlas of age-associated transcriptomic alterations across human tissues." Communications Biology (2025). https://www.nature.com/articles/s42003-025-08845-8
5. Luecken, M. D. and Theis, F. J. "Current best practices in single-cell RNA-seq analysis: a tutorial." Molecular Systems Biology (2019). https://link.springer.com/article/10.15252/msb.20188746
6. Luecken, M. D. et al. "Benchmarking atlas-level data integration in single-cell genomics." Nature Methods (2022). https://www.nature.com/articles/s41592-021-01336-8
7. Uyar, B. et al. "Single-cell analyses of aging, inflammation and senescence." Ageing Research Reviews (2020). https://pmc.ncbi.nlm.nih.gov/articles/PMC7493798/
8. Jin, K. et al. "Brain-wide cell-type-specific transcriptomic signatures of healthy ageing in mice." Nature (2025). https://www.nature.com/articles/s41586-024-08350-8
9. Denisenko, E. et al. "Systematic assessment of tissue dissociation and storage biases in single-cell and single-nucleus RNA-seq workflows." Genome Biology (2020). https://pubmed.ncbi.nlm.nih.gov/32487174/