Human and Mammalian Brain Atlas Release: Basal Ganglia

The single-cell transcriptomics data that was clustered to generate this taxonomy is available for visualization, programmatic use, and download.

The consensus basal ganglia cell type taxonomy is incorporated directly into the Cell Ontology (pre-print) - the community standard for cell type annotation used by major single cell data portals such as CZ CELLxGENE and the Broad Single cell Portal. Integration is semi-automated and ensures that the Cell Ontology reflects the taxonomy hierarchy. All HMBA cell types in CL and their markers are linked directly to reference data on the Allen Brain Cell Atlas. All types are linked to locations defined in the Harmonized Ontology of Mammalian Brain Anatomy (HOMBA). Integration into CL also serves to integrate cell types with other standards, including the types defined by the human Cell Atlas and to map types to HMBA defined types to classically defined cell types and their synonyms.

The Harmonized Ontology of Mammalian Brain Anatomy (HOMBA) is a harmonized cross-species taxonomy of 2341 brain and spinal cord structures. Derived from the Allen Developing Human Brain Atlas (DHBA) ontology, the HOMBA is hierarchical, allowing users to aggregate structures from fine grain parcellations to broad regions. Terminology is harmonized across human, primate, and rodent structures with synonymous terms and includes transient developmental structures. HOMBA is designed for neuroanatomical applications including brain sampling and dissection, tissue block mapping, atlas building, cell-type and pathology localization, and linking cross-species and developmental datasets.

The Common Coordinate Framework Multispecies Atlas Primer (CCF-MAP) is a resource for accessing marmoset, macaque, and human atlases developed by the Human and Mammalian Brain Atlas (HMBA) consortium, as well as the CCFv3 mouse atlas from the Allen Institute. CCF-MAP is a repository of publicly accessible MRI templates, associated parcellations of basal ganglia structures, and related metadata, which can be viewed and downloaded. CCF-MAP also includes Jupyter notebook tutorials for querying voxel data and for downloading region masks for offline use.

MapMyCells allows you to assign cell type names from Allen Institute-hosted taxonomies to your own single cell and spatial transcriptomics data. The Mammalian Basal Ganglia Consensus Cell Type Taxonomy is now available as a mappable taxonomy in MapMyCells. Our mapping algorithm now includes species detection and automatic ortholog conversion.

An example notebook is provided showing how to map unlabeled data onto the taxonomy using the Python version of MapMyCells. At this point, files are separated by species. Human data should be mapped with the human data files, macaque data with the macaque data files, and marmoset data with the marmoset data files.

Human: precomputed_stats in AWS S3 --- query_ markers in AWS S3
‍Macaque: precomputed_stats in AWS S3 --- query_ markers in AWS S3
Marmoset: precomputed_stats in AWS S3 --- query_ markers in AWS S3

The Allen Brain Cell (ABC) Atlas Access package provides methods for downloading the HMBA-BG data and metadata programmatically via python. The repository also provides Jupyter notebooks showing examples of how to join and utilize these data products, allowing users to delve deeper into visualizations and results they create with the ABC Atlas visualization tool.

ATAC-seq data are available at various stages of processing with snaptac2: (1) fragment files in .h5ad format, (2) Group level bigwigs and (3) MACS3 peak calling. Data for all species and levels of processing can be found at the public AWS s3 bucket: s3://hmba-bican-sharing-802451596237-us-west-2/BasalGanglia_pre-print_ATAC/.

This release also includes more than 1,000 patch-seq cells, sampled from across the Basal Ganglia in macaque. Multimodal data files, cell features, and cell type labels from the consensus taxonomy will soon be available for download. Data can be visualized in the Cytosplore Viewer as well. (See Tools section above)

To unify the field’s understanding of basal ganglia cell types across species, efforts supported by the NIH BRAIN Initiative have initiated a comprehensive nomenclature system that synthesizes pre-existing literature describing cell type nomenclature with new cross-species transcriptomic data. This effort includes coordinated contributions from the Human and Mammalian Brain Atlas (HMBA) project in the BRAIN Initiative Cell Atlas Network (BICAN), and the Basal Ganglia AAV Toolbox project of the BRAIN Initiative Armamentarium for Precision Cell Access. This initiative integrates HMBA single-nucleus RNA sequencing (snRNA-seq) data from human, macaque, marmoset, and previously published mouse basal ganglia, with the goal of generating a consensus cell type taxonomy that can be widely adopted by the scientific community. By focusing on conserved marker genes and shared molecular profiles to supplement established names from the broader community, we have developed a standardized naming system that captures the evolutionary relationships and functional distinctions among basal ganglia cell types. The HMBA consensus basal ganglia taxonomy is designed to streamline communication, foster collaboration, and facilitate the development of novel research tools targeting specific cell types across multiple species.

Our consensus basal ganglia cell type taxonomy is the result of iterative clustering and cross-species integration of transcriptomic data from single-nucleus 10x Genomics multiomic profiling. The taxonomy encompasses neurons from key structures within the basal ganglia, including the caudate (Ca), putamen (Pu), nucleus accumbens (NAc), the external and internal segments of the globus pallidus (GPe, GPi), ventral pallidus (VeP), subthalamic nucleus (STN), and substantia nigra (SN). By combining data from multiple primate and rodent species, we have developed a consensus taxonomy that highlights both conserved and species-specific cell types. We validate our taxonomy through marker gene expression analysis, comparison with previously published taxonomies, and self-projection, ensuring the accuracy and robustness of each level in the taxonomic hierarchy.

To support the HMBA consensus cell type taxonomy, we have compiled extensive metadata providing detailed information associated with each identified cell type. These metadata include information on gene expression patterns and marker genes from snRNA-seq as well as synonymous names from existing literature, including Yao et al. 2023 and Siletti et al 2023. The metadata fields will be organized under an aligned taxonomy data format enabling integration with pre-existing tools such as CellxGene and Cell Annotation Platform.

In collaboration with the BRAIN Armamentarium Consortium, we have developed an extensive collection of cell type-specific viral tools, designed to target and manipulate specific cell types identified in our cross-species integrated consensus BG taxonomy (Hunker, Wirthlin, et al. 2024). Using bulk or single nucleus ATAC-seq and multiomic data, we identify cell type-specific gene regulatory elements—particularly enhancers—linked to the cell types we identified through snRNA-seq. These efforts have culminated in a robust computational pipeline for identifying successful enhancers with a high degree of accuracy, the Cross-species Enhancer Ranking Pipeline (CERP). This pipeline allows us to identify and prioritize testing of the most promising candidate cell type enhancers for targeting homologous cell types across mammalian species. This viral tool collection comprises hundreds of enhancer adeno-associated virus (AAV) vectors, each with detailed brain expression analysis catalogues in the Genetic Tools Atlas. The most promising tools were also subjected to molecular validation. Additionally, we offer variants of these tools with enhancers optimized for stronger or more specific labeling, as well as variants delivering fluorescent proteins, channelrhodopsin variants, GCaMP variants, recombinases (Cre and Flp) to enable targeted experimental manipulation for functional studies. Many of these enhancer AAV vectors are already available through Addgene, with new tools continuously being added to the collection. These collective resources provide researchers with a powerful toolkit for targeting and studying the diverse cell types of the basal ganglia across multiple species to advance our understanding of BG function and dysfunction in health and disease.

Cell type subclasses in BG regions are conserved between mouse, human, and non-human primates. Candidate enhancers were identified for the major BG cell subclasses using a novel computational pipeline we call Cross-species Enhancer Ranking Pipeline (CERP), which examines open chromatin using single nucleus ATAC-seq datasets. Additional enhancers were selected using CERP to target other BG cell populations, including groups of cell types with regional enrichment, such as dorsal vs. ventral striatum.
‍