Spatial atlas of 8 million cells maps cell types in the human basal ganglia

Researchers at the University of California, Irvine have produced a spatial atlas of roughly 8 million cells from the human basal ganglia, a set of deep brain structures involved in movement, motivation, habits and decision-making. The map identifies 60 distinct cell types and shows how they group into local cellular communities. The study, published in Cell, is intended as a reference for work on conditions such as Parkinson's disease and Huntington's disease.

Adding location to cell identity

Single-cell sequencing has made it possible to distinguish brain cell types by their gene activity. The approach usually requires cells to be separated from the tissue, which discards information about where they sat and what surrounded them. Xiangmin Xu, who directs UC Irvine's Center for Neural Circuit Mapping, argues that location and neighbourhood are as important as the catalogue of types itself.

The team therefore used spatial transcriptomics, which measures gene activity while keeping cells in place within the tissue. Two technologies, MERFISH+ and Stereo-seq, were combined and applied to brain samples from four neurotypical donors. This allowed the researchers to link molecular cell identities directly to the anatomy of the basal ganglia.

What the map shows

Different cell types were found to occupy distinct areas of the basal ganglia. Some supporting cells gathered along the borders between compartments. Certain kinds of neurons that are more abundant in primates were confined to particular compartments.

Xu describes the analysis of cellular communities, meaning the combinations of cells that occur together in specific locations, as a central advance. Such relationships cannot be read from a list of cell types alone. The first authors named among the major contributors are Bereket Tesfay Berackey, Zhiqun Tan, Ginny Wu and Sujan Das, with Bogdan Bintu of UC San Diego a key collaborator on technology development, experimental design and data analysis.

A baseline for disease research

The atlas gives researchers a framework for comparing tissue from people with and without brain disorders. One possible use is to ask why some cell populations are vulnerable in disease while others are relatively resistant, and whether neighbouring cells help explain the difference. Over time, such work could inform more precise experimental models and the identification of therapeutic targets.

The source is clear that the atlas is a foundational resource, and that any clinical value depends on later studies. The reference is also built from four donors without neurological disease, so comparisons with disease tissue remain a task for future work.

Infrastructure behind the data

A companion paper in Cell Systems, led by Jillian V. Berry, Bryan M. Tran, P. Adolfo Sequeira and Lorena Valenzuela, describes UC Irvine's brain procurement programme. It sets out workflows that link donor identification and tissue recovery to the quality standards and anatomical documentation needed for advanced molecular work. The programme operates with partners including the Orange County Coroner's Office, UC Irvine's Willed Body Program, UCI Health, CHOC and OneLegacy. The authors also note difficulties in recruiting donors who reflect the population of Southern California, and point to opportunities to widen outreach and donation pathways.

Part of a wider consortium effort

Both studies belong to a coordinated publication package from the BRAIN Initiative Cell Atlas Network (BICAN), launched in 2022. The package involves hundreds of scientists at 55 institutions in nine countries and proposes a shared framework for classifying and comparing basal ganglia cells across species. Common classifications should help laboratories reconcile their findings and judge how results from animal models relate to human biology. The methods and standards are also meant to guide mapping of the rest of the brain.

Xu likens the long-term potential of BICAN to that of the Human Genome Project as a common reference, while noting that a brain atlas must integrate many kinds of molecular data with the arrangement of cells and circuits. The next step at the Center for Neural Circuit Mapping is to connect these molecular maps with data on anatomical connections, cell activity and disease-related change.

Source: Molecular mapping of 8 million cells brings cerebral architecture into focus (Medical Xpress, 13 November 2026).

Image: AI-generated illustration.