Delattre, E. | Search bioRxiv

bioRxiv2024

An extended and improved CCFv3 annotation and Nissl atlas of the entire mouse brain

Brain atlases are essential for quantifying cellular composition in mouse brain regions. The Allen Institutes Common Coordinate Framework version 3 (CCFv3) is widely used, delineating over 600 anatomical regions, but it lacks coverage for the most rostral and caudal brain parts, including the main olfactory bulb, cerebellum, and medulla. Additionally, the CCFv3 omits key cerebellar layers, and its corresponding Nissl-stained reference volume is not precisely aligned, limiting its utilisability. To address these issues, we developed an extended atlas, the Blue Brain Project CCFv3 augmented (CCFv3aBBP), which includes a fully annotated mouse brain and an improved Nissl reference aligned in the CCFv3. This enhanced atlas also features the central nervous system annotation (CCFv3cBBP). Using this resource, we aligned 734 Nissl-stained brains to produce an average Nissl template, enabling an updated distribution of neuronal soma positions. These data are available as an open-source resource, broadening applications such as improved alignment precision, cell type mapping, and multimodal data integration.

bioinformatics↗

bioRxiv2024

Deep learning for classifying neuronal morphologies: combining topological data analysis and graph neural networks

Neuronal shape determines how neurons process and integrate information, yet a consistent and objective classification of neuronal morphologies remains elusive. Current approaches rely heavily on subjective expert views or on predefined features, limiting reproducibility and interpretability. Here, we present an interpretable deep learning framework that unifies topological data analysis, graph neural networks, and traditional morphometrics to classify neuronal morphologies objectively and transparently. Our framework compares complementary mathematical representations of neurons to capture geometric, topological, and graph-structural information. Then it benchmarks their performance against expert-labeled datasets. We show that topology- and graph-based models achieve accuracies comparable to human experts, revealing that both global branching invariants and local connectivity patterns are essential to define morphological cell types. Using explainable artificial intelligence methods, we identify structural features driving each classification decision, bridging computational and neuroanatomical interpretations. This open source and reproducible approach provides a foundation for scalable, interpretable and biologically meaningful neuronal taxonomy, enabling consistent comparisons between data sets and species.

neuroscience↗