Using the BSB, the modelling team of Pavia has modelled the intricate geometry of the cerebellar cortex and simulated its electrical dynamics. The model has been used to investigate pathologies and fundamental properties of single cells operating in the network, spanning all the way down to subcellular perturbations of biological mechanisms and up to the network level. This has allowed to successfully address short- and long-term plasticity changes and to generate models with a higher level of abstraction, including spiking neural networks and mean field models. These, once integrated into The Virtual Brain, are allowing to investigate the causal relationship between elementary causes and ensemble signals of clinical relevance, such as BOLD and LFP. Furthermore, the tool is being used for student training, to provide practical experience with coding, modelling, simulations and result analysis.
Brain Scaffold Builder
Flexible and organized workflows for neural circuit reconstruction and simulation at different levels of detail
Use the Brain Scaffold Builder (BSB) to reconstruct realistic neural circuits by placing and connecting fibres and neurons with detailed morphologies or only simplified geometrical features. Configure your model the way you need. Interfaces with several simulators (NEURON, Arbor, NEST) allow simulation of the reconstructed network and investigation of the structure-function-dynamics relationships at different levels of resolution. The “scaffold” design allows an easy model reconfiguration reflecting variants across brain regions, animal species and physio-pathological conditions without dismounting the basic network structure. The BSB provides effortless parallel computing both for the reconstruction and simulation phase.
- Define network topologies (using mesh, 3D-voxel, and atlas support) and process neuron morphologies using our validation and transformation libraries
- Place and orient neurons and fibres using multiple placement strategies, and locate synapses using multiple connectivity strategies (e.g., touch detection, voxel intersection, distance-based probability functions)
- Pick-up or design de novo interchangeable blocks to create advanced modelling workflows in a user-friendly framework
- Define the detail level of simulations (e.g., multicompartmental neurons or point neurons) targeting a specific simulator backend
- Describe the neuron and synapse properties in a user-friendly framework
- Introduce alterations, by simple re-parameterization in the configuration file, to simulate the spatio-temporal propagation of lesions or modifications at network level
Get started
Get started
Get started: Explore our notebooks to place neurons by matching experimental density distributions, in any arbitrary volume or in regions extracted from the Allen Mouse Brain Atlas CCFv3. Assign synapses based on touch detection, voxel intersection, distance-based probability functions or your customized approach. Define dynamic features of each neuron and synapse, stimulation protocols, and recorded variables, in the configuration file. For complete information you can read through our documentation.
Install the BSB using pip, or use it on EBRAINS’ Jupyter Lab, and get started right away!
User story
Future perspectives
Stay tuned for an ever-expanding list of features and integrations! We have exciting new expansions planned for model optimization with BluePyOpt, automatic LFP generation with LFPy, morphology synthesis, generative networks, and whole-brain atlas reconstructions. With every new release we improve efficiency and integrate new tools. We have already integrated the most common tools and data sources for computational neuroscience, e.g NeuroMorpho, Allen Mouse Brain Atlas, NEURON, Arbor, NEST, MorphIO, SONATA, Neo. The BSB is therefore going to be a pivotal element in the multiscale modelling workflow of EBRAINS.
Related publications
De Schepper R, Geminiani A, Masoli S, Rizza MF, Antonietti A, Casellato C and D’Angelo E. Model simulations unveil the structure-function-dynamics relationship of the cerebellar cortical microcircuit. NAT COMMS BIOL. 2022; doi: 10.1038/s42003-022-04213-y