Atlas-based Reconstruction of the Mouse Cerebellar Flocculus

Brain modelling is a promising approach to investigate the structure and the dynamics of the brain, both in physiological and pathological conditions, and integrate knowledge across various scales. Bottom-up models can incorporate detailed biological features into large-scale network with a data-driven approach. These offer an accurate representation of the corresponding natural circuits. In this Thesis’ Project a large-scale network of the mouse cerebellar Flocculus has been reconstructed in the context of developing models with regional specificities. The network volume is mapped on the real shape of the Flocculus (~1,4 mm3), and it derives from the Allen Mouse Brain Atlas. A new type of neuron excluded from previous canonical non-region-specific cerebellar scaffold models is introduced: the Unipolar Brush Cells. This cell type has a particularly high density in the Floccular region and is important to modulate the input signal in the Granular Layer. The different cell densities are derived from the 3D Blue Brain Mouse Cell Atlas version 2, correlating literature data with neuron excitatory and inhibitory densities. The final network comprised the Granule, Golgi, Unipolar Brush, Purkinje, Basket, Stellate cells and the mossy fibers innervation (for a total of ~1.3 million elements), distributed in the Granular, Purkinje and Molecular layer of the cerebellar cortex. The placement and the connectivity were computed with the Brain Scaffold Builder, a component framework for neural modelling developed at the Department of the Brain and Behavioral Sciences of the University of Pavia, exploiting supercomputers for computations. The Flocculus model obtained with this work is being simulated as a point-neuron spiking network with the NEST simulator, yielding to a comprehensive understanding of the dynamics of the circuit. The entire workflow relies on experimental data. Hence, when more data will become available, the model can be updated to reflect the current state of knowledge on cell distribution and connectivity. The pipeline employed to reconstruct the Flocculus is flexible and can be adapted to reconstruct other brain regions or expanded to other parts of the olivocerebellar system. As soon as computational resources became available, it could be generalized to obtain a computational model of the whole brain.