Many excellent brain atlases exist for different species. Some of them have an API (application programming interface) to allow users to interact with the data programmatically (e.g. the excellent Allen Mouse Brain Atlas), but many do not, and there is no consistent way to process data from multiple sources.
The brainglobe atlas API (BG-AtlasAPI) deals with this problem by providing a common interface for programmers to download and process data from multiple sources.
Each atlas consists of data files in a common format:
A "reference" image of a brain (.tiff
)
An "annotation" image, with each brain region defined by a unique pixel value (.tiff
)
Meshes defining the surface of each brain region (.obj
)
A mapping of brain region pixel value to region name, and structure hierarchy (.json
)
Metadata defining the shape, orientation etc. of the data, and other info such as animal species and authors (.json
)
A number of atlases are in development, but those available currently are:
The Allen Mouse Brain Atlas at 10, 25 and 50 micron resolutions
The Allen Human Brain Atlas at 100 micron resolution
The Max Planck Zebrafish Brain Atlas at 1 micron resolution
The Enhanced and Unified Mouse Brain Atlas at 25 micron resolution
BG-AtlasAPI works with Python >3.6, and can be installed from PyPI with:
pip install bg-atlasapi