Behold the mouse brain, in glorious new detail. In the May 14 Cell, researchers led by Julie Harris and Lydia Ng at the Allen Institute for Brain Science in Seattle debut the third iteration of the Allen Mouse Brain atlas. Called the Common Coordinate Framework version 3 (CCFv3), this latest version takes advantage of high-resolution imaging techniques to produce a three-dimensional map of the entire brain at cellular-level anatomical detail. It also incorporates multiple types of gene expression, protein, and connectivity data to delineate each brain subregion. Researchers can import their own data into the atlas and use its common reference framework to compare their findings with those of other labs, Harris said.

  • The Allen Institute has released the third version of its mouse brain atlas.
  • It is now three-dimensional, and 1,000-fold higher resolution than the previous version.
  • It incorporates gene, protein, and connectivity data to define each subregion.

“This three-dimensional reference atlas is a true tour de force that will be hugely useful for the neuroscience community,” Tara Spires-Jones at the University of Edinburgh wrote to Alzforum (full comment below).

The original Allen Reference Atlas (ARA), released in 2008, consisted of two-dimensional coronal and sagittal sections spaced every 200 μm through the brain. Version 2, in 2011, converted this atlas into three dimensions by brute force, extrapolating what might lie between each section to achieve a resolution of 100 μm.

See It Clearly. The third Allen Mouse Brain atlas used high-resolution imaging to draw a detailed three-dimensional map of brain structure. [Courtesy of Wang et al., Cell.]

For version 3, however, Allen researchers made use of serial two-photon tomography to map the entire mouse brain to a resolution of 10 μm, 1,000-fold higher than the previous version. This resolution is in the range of neuronal cell bodies in the mouse brain, which vary in width from five to 20 μm. The researchers imaged 1,675 young adult C57BL/6J mice and mathematically averaged the results into a single template. To their surprise, this iterative process heightened anatomical details, revealing fine features such as barrel formations in the somatosensory cortex (see "The Power of Averages" below). They were able to map the true three-dimensional geometry of the cortex, including the thick dendrites of layer V pyramidal neurons (see movie).

Joint first authors Quanxin Wang, Song-Lin Ding, and Yang Li then overlaid data from histology, immunohistochemistry, in situ hybridizations, transgene expression, and connectivity tracing to define brain subregions. They used nomenclature from the ARA and another classic two-dimensional mouse atlas, Paxinos and Franklin’s Mouse Brain in Stereotaxic Coordinates. This approach resulted in 658 individual named brain structures (see "A Brain Divided" below).

The Power of Averages. Averaging 1,675 mouse brains brought out fine structural details. [Courtesy of Wang et al., Cell.]

A Brain Divided. Multiple types of brain data allowed for the delineation of individual subregions. [Courtesy of Wang et al., Cell.]

All the data are freely available on the Allen Institute website. Harris encourages the research community to point out any discrepancies with existing data and submit new results to help refine and update the atlas.

“This will no doubt be a useful teaching tool, in addition to its great utility to the research community,” Michael Sasner at the Jackson Laboratory in Bar Harbor, Maine, wrote to Alzforum (full comment below). He collaborates with Harris on another project.—Madolyn Bowman Rogers

Comments

  1. This three-dimensional reference atlas is a true tour de force that will be hugely useful for the neuroscience community and for AD researchers working with mouse models on a C57Bl6/J background. One example is the detailed projection data will be very useful for studies of pathological protein propagation through neural circuits. In future, it would be amazing if this atlas could get to even higher resolution to integrate synaptic structure and connectivity data.

  2. The mouse brain is the most widely used model in modern neuroscience. This essential new paper describes the latest release (v3) of the Allen Mouse Brain Common Coordinate Framework (CCF). This is based on averaging images from 1,675 genetically identical mouse brains and various datasets from histological staining, immunohistochemistry, analysis of transgenic expression, in situ hybridization, and tracer connectivity analysis. Integration of all this data results in a cellular-level resolution three-dimensional mouse reference brain, with delineated gray-matter structures, fiber tracts, and ventricular structures.

     As both basic and translational research continue to move toward combining various types of analysis at different spatial resolutions (from functional studies to single-cell transcriptomics), tools that enable integration of these datasets become more and more essential. The CCF provides the dataset and resources needed for this integration.

    Perhaps most critically, kudos to the Allen Institute for Brain Science team for being a model of open science. They have enabled unrestricted access to over 100 mouse lines that express cre in specific neuronal populations that were used to help define brain structures in this paper, as well as dozens of cre reporter lines. The data described here are available for download, as well as accessible via a user-friendly interactive Atlas Viewer and supported by an active, well-supported online community forum. This will no doubt be a useful teaching tool, in addition to its great utility to the research community. 

    Full disclosure: I have a research collaboration with one of the lead authors.

Make a Comment

To make a comment you must login or register.

References

Basic page Citations

  1. The Allen Mouse Brain Common Coordinate Framework

External Citations

  1. CCFv3
  2. Mouse Brain in Stereotaxic Coordinates

Further Reading

Primary Papers

  1. . The Allen Mouse Brain Common Coordinate Framework: A 3D Reference Atlas. Cell. 2020 Apr 30; PubMed.