Brain Atlas Maps Epigenetic Changes Associated with Aging in Mice

Why It Matters

By providing a high‑resolution reference of how the brain epigenome remodels with age, the atlas enables faster identification of molecular drivers of neurodegeneration and supports development of targeted therapeutics. Its open access accelerates cross‑species translation for aging research.

Key Takeaways

• Atlas covers 8 brain regions, 36 cell types
• Over 200k cells profiled for methylation and chromatin
• Non-neuronal cells show strongest age‑related methylation loss
• TAD boundary strength identified as aging biomarker
• Spatial data reveals region‑specific inflammation patterns

Pulse Analysis

Epigenetic alterations are recognized as a central driver of brain aging, yet the field has lacked a unified, cell‑type‑resolved map to connect molecular changes with functional decline. Traditional bulk analyses mask the heterogeneity of neuronal and glial populations, making it difficult to pinpoint which cells are most vulnerable. The new Salk Institute atlas fills this gap by delivering single‑cell resolution across methylation, chromatin architecture, and spatial transcriptomics, offering researchers a granular view of the aging brain that was previously unattainable.

The resource spans eight mouse brain regions and 36 distinct cell types, encompassing more than 200,000 cells profiled for DNA methylation and three‑dimensional genome organization, plus nearly 900,000 cells captured with spatial transcriptomics. Key discoveries include a pronounced loss of methylation at transposable elements in non‑neuronal cells, and the emergence of stronger topologically associating domain (TAD) boundaries and heightened CTCF accessibility as robust aging biomarkers. By linking these epigenetic signatures to region‑specific inflammation patterns, the atlas clarifies how local microenvironments modulate cellular senescence.

Because the dataset is openly available on Amazon Web Services and the Gene Expression Omnibus, it can be directly integrated with human brain atlases and the NIH BRAIN Initiative’s multimodal collections. Early adopters have already leveraged the data to train deep‑learning models that predict age‑related gene expression shifts, accelerating hypothesis generation for neurodegenerative disease mechanisms. As the community builds on this reference, it is poised to guide target validation, biomarker discovery, and precision‑medicine strategies aimed at mitigating age‑associated cognitive decline.