The Brain Scaffold Builder

Developed by the Department of Brain and Behavioral Sciences at the University of Pavia, the Brain Scaffold Builder ( BSB) is a component framework for neural modeling. It emphasizes modular component declarations to assemble brain models, supporting various configuration languages and Python library functions. BSB facilitates parallel reconstruction and simulation of diverse network topologies, placements, and connectivity strategies.

Monorepo Structure

The BSB suite is now organized as a monorepo, integrating the following components:

• bsb-core: The foundational framework of BSB.
• bsb-hdf5: Storage engine for saving models in HDF5 format.
• bsb-json: Parser for reading and writing JSON configuration files.
• bsb-yaml: Parser for reading and writing YAML configuration files.
• bsb-nest: Simulation interface for point-neuron networks using the NEST simulator.
• bsb-neuron: Simulation interface for detailed neuron networks using the NEURON simulator.
• bsb-arbor: Simulation interface for detailed neuron networks using the ARBOR simulator.
• nmodl-glia: An NMODL asset manager.
• nrn-patch: Utility layer to reduce friction when working with NEURON in Python.
• arborize: Tool for Arbor-like descriptions of cell models for NEURON and Arbor.
• bsb-test: Tools and configurations for testing BSB components.
• sphinxext-bsb: Sphinx extension for BSB documentation.

Documentation

Comprehensive documentation is available at:

• BSB Documentation

Installation

Pip installation

The BSB supports Python 3.10 and above. To install the BSB suite along with the core framework and default plugins:

pip install bsb

For advanced users desiring a custom combination of plugins, install bsb-core and the desired plugins separately:

pip install bsb-core
pip install bsb-hdf5 bsb-json bsb-yaml

To include simulation support for NEST, NEURON, or ARBOR:

pip install bsb[nest]
pip install bsb[neuron]
pip install bsb[arbor]

Clone installation

You can also clone the repository and use the devtools script to install all packages in editable mode in your own environment:

git clone https://github.com/dbbs-lab/bsb
cd bsb
pip install -r devtools/editable-install.txt

Developer installation

For development purposes, you should install the UV and NX tools. To this end, there is a helper script for each platform in devtools/bootstrap-*. For instance, for linux:

git clone https://github.com/dbbs-lab/bsb
cd bsb
./devtools/bootstrap-linux.sh

From there, you NX and UV will automatically create an environment for each subpackage, including all necessary libraries for running the unittests, linting the code and building the documentation.

For more information, please refer to the developers' documentation.

Usage

To create a new BSB project:

bsb new my_model --quickstart
cd my_model

This command generates a my_model directory with starter files:

• network_configuration.yaml: Defines the network configuration.
• pyproject.toml: Contains project metadata and configuration.
• placement.py and connectome.py: Custom components for network placement and connectivity.

To compile the network:

bsb compile

For simulation, refer to the documentation for configuring simulations with NEST, NEURON, or ARBOR.

Contributing

All contributions are very much welcome. Take a look at the contribution guide

Acknowledgements

This research has received funding from the European Union’s Horizon 2020 Framework Program for Research and Innovation under the Specific Grant Agreements No. 945539 (Human Brain Project SGA3) and No. 785907 (Human Brain Project SGA2), as well as from the Centro Fermi project “Local Neuronal Microcircuits” to ED. The project also receives funding from the Virtual Brain Twin Project under the European Union's Horizon Europe program under grant agreement No. 101137289.

We acknowledge the use of the EBRAINS platform and Fenix Infrastructure resources, which are partially funded by the European Union’s Horizon 2020 research and innovation program under the Specific Grant Agreement No. 101147319 (EBRAINS 2.0 Project) and through the ICEI project under grant agreement No. 800858, respectively.

Supported by