107 — An interactive framework for whole-brain maps at cellular resolution

Most modern-day connectomics projects aim to collect as large an anatomical image space as possible without sacrificing resolution. This project, leveraging brain atlases created by the Allen Brain Institute, collects many imaging modalities such as fluorescence data, and overlays this imagery on the map to improve our understanding of connectivity both within brain regions and across cortex.

The big contribution (at least as I see it) is that the authors of this paper provide their software as a public resource in order to enable others to leverage the same annotation and visualization systems they used for publication. This sort of consistency is hugely important for science reproducibility between collaborators and other research laboratories.

The paper makes the distinction between vector-based and raster-based representations of neuroanatomy. Their online demo leverages vector-based imagery to render maps at high resolution: By considering the annotations and anatomical features of a brain as vector-based shapes, the visualization software is able to represent structures at any level of resolution, limited only by the resolution of the imagery or annotation itself.

In addition, this paper demonstrates a pipeline for coregistering brains with the Allen (mouse) atlas and then tracing fiber tracts across the volume of the brain in order to approximate connectivity patterns, or trace cell gene expression, or visualize other patterns in brain-space.

I hope that this work will continue to inspire other researchers to contribute to these open-source neuroscience research projects and begin to collaborate on industry-standards for neuroimagery analysis. The sooner we can begin to agree on image-spaces, the sooner whole-brain collaborations will become possible.