Reaching for a coffee, repeating a habit, deciding to move: each passes through a cluster of cells at the base of the forebrain called the basal ganglia. On Nov. 12, the National Institutes of Health announced that researchers in its BRAIN Initiative had, for the first time, produced a comprehensive map of that region, cataloguing the activities, identities and locations of millions of individual cells. The atlas covers about 25.5 million cells drawn from humans, macaques and marmosets.
Read next on Morning Overview: 11 engines built to run well past 200,000 miles
NIH’s release calls the basal ganglia a region critically important for turning thoughts into actions and nicknames it the brain’s behavioral gatekeeper.
A 25.5 million-cell atlas across three primates
The NIH release introduces the atlas as the work of the BRAIN Initiative Cell Atlas Network, known as BICAN. Reports from the Allen Institute’s side of the project put the total at 25.5 million cells, roughly 21 million human, 3 million macaque and 1.5 million marmoset, according to Nature World News. The Scientist adds that the team used single-nucleus transcriptomics, supplemented with epigenomic, spatial and multiomic data, and that the findings were published as a package of studies in Cell Press journals.
The donor base is small for the human portion: the human cells came from six brains, the macaque cells from three animals, and the marmoset cells from two, while the consensus taxonomy was drawn from a cross-species dataset of about 1.8 million nuclei. Hongkui Zeng of the Allen Institute called it a major landmark reference atlas. Her colleague Ed Lein explained the choice of region: the basal ganglia made a good starting point for a cell atlas, in his telling.
The consortium’s shared naming system is part of the product. Tech Times reports a consensus taxonomy of 61 cell groups, built by more than 600 members at 55 institutions in nine countries. Evan Macosko of Harvard Medical School and the Broad Institute said labs lacking a uniform naming system struggle to compare notes, and the consensus labels, released through the Allen Institute’s Human and Mammalian Brain Atlas pages, are meant to fix that.
Cell types tied to schizophrenia and autism
The region is not one block of tissue. Tech Times lists its deep-brain structures as the striatum, the globus pallidus, the subthalamic nucleus and the substantia nigra, which help start and sequence voluntary movement and also contribute to learning, emotion and cognition. The atlas matters to disease researchers because this region is implicated in Parkinson’s disease, Huntington’s disease and obsessive-compulsive disorder. The Scientist reports that one group found a previously unrecognized type of spiny projection neuron carrying a serotonin receptor, and that its presence was associated with elevated schizophrenia risk. A second group reported a new cell type linked to schizophrenia and bipolar disorder, and a third found genetic changes in certain basal ganglia cells associated with autism.
Cell organization into domains also appears largely conserved across the three primate species, which suggests that findings in macaques and marmosets can inform human biology.
Fenna Krienen of Princeton, per Tech Times, said spatial context can overturn textbook definitions of what makes a cell type, and she likened the effort to an early draft of a genome project. Macosko described the striatum’s layout as the first clear evidence of an architecture to that structure. Tech Times notes that he is also an academic co-founder of Curio Bioscience, which licenses spatial-genomics technology, and that Princeton’s Ning Wang built a deep-learning tool called PIANO that is now used across the consortium. The datasets and mapping tools, including MapMyCells and the ABC Atlas browser, are public, and the Allen Institute’s release page says species-specific volumes, cortical surface representations and spatial data aligned to MRI will follow in later releases.
A parts list built from six human brains
The limits are laid out in the same coverage. The human reference rests on cells from six brains, so differences between people need larger cohorts, and the atlas is a parts list that does not show how each cell type shapes behavior. Spatial panels used up to 300 genes, and only 72 were shared across all three species. Peter Strick of the University of Pittsburgh, who was not involved, called the atlas a major first step and said the next job is working out how subtypes connect to one another and to areas such as the cortex.
The gene therapy result that reduced tremors in one project was tested only in animal models. Zeng said the next taxonomy version will not be the definitive one, and she hopes a whole-brain primate atlas could be finished within five years. Joseph Ecker of the Salk Institute pointed to AI models trained on atlas data as the next step, for example to predict how a disease-linked genetic variant affects specific neurons.
Money is the open question. Tech Times reports that Cures Act funding ends after fiscal 2026, that BICAN is scheduled to run through 2027 with a possible extension to 2028, and that an NIH BRAIN Initiative official said the program is looking for ways to fund a whole-brain atlas after the network ends.
This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.
More from Morning Overview
- A $1 billion gas plant planned for Superior, Wisconsin, has been abandoned
- Shaq spent $1.3 million on three Rolls-Royce Phantoms to prove a point to a salesman
- The Justice Department asked courts to undo approvals for three offshore wind farms, one already making power
- Over 20 trillion gallons of rain fell on the Four Corners states and Nevada in two weeks