The Brain Image Library: New features for enhanced data integration, accessibility, and visualization

The Brain Image Library (BIL) is the centralized repository of brain microscopy data. BIL provides access to a diverse range of whole-brain microscopy image datasets, along with secondary data such as neuron morphologies, connectivity between cells, spatial transcriptomics, and historical collections of significant value to the community. BIL welcomes contributions of microscopy data relevant to the BRAIN Initiative, including data from primates, most mammals, and model organisms.

One core mission of the Brain Image Library is to utilize available computing resources to make data easily visualizable. We are developing an on-demand transformer, BrAinPI, to optimize the accessibility and reuse of the diverse datasets at BIL. This will facilitate the delivery of data in various formats and provide flexibility to the neuroscience community. This tool is already serving our Neuroglancer data viewer, and is compatible with visualization tools like Napari or Neuroglancer. Using the Neuroglancer data viewer at BIL, users can visualize petascale datasets even on a smartphone, at the speed of mobile internet, without registration or downloading data. Additionally, we are developing a way to easily visualize 2D histology data in a browser as high-resolution zoomable images using a lightweight OpenSeaDragon viewer. Lastly, we have developed a plugin for napari called napari-bil-data-viewer that enables users to visualize publicly available datasets at BIL over the web without registration and is pre-loaded with a comprehensive set of downsampled whole brain fMOST datasets and associated neuron morphology SWC files.

Here, we will also present new metadata features at BIL including a new metadata schema for spatial transcriptomics data to increase the reusability of such data and provide a better description of these data in BIL. We have also developed a Metadata API and web portal that enables full-text searching or searching of individual metadata fields. These tools significantly enhance the accessibility, findability, and reusability of data at BIL. As a whole, BIL’s computational, visualization, and metadata tools aim to empower the neuroscience community by providing a user-friendly platform for utilizing a wide range of imaging datasets.