Dataset and Pre-computed Matrices for the Neuro-Informed Generative Connectome (NIGC) Framework

Overview This dataset serves as the supplementary data repository for the GitHub codebase of the Neuro-Informed Generative Connectome (NIGC) framework. Due to GitHub's file size limitations, the raw 3D neuronal coordinate data and the large pre-computed intermediate matrices generated during our experiments are hosted here. These files are essential for fully reproducing the structural and functional analyses presented in our manuscript without needing to re-execute extremely time-consuming computational steps.
File Descriptions
1. NIGC_microdata.zip
Contents: This archive contains the microdata directory, which includes raw 3D spatial coordinate files (.txt) for neurons across various mouse brain regions (e.g., ACx, HPC, MGB, etc.).Source: The coordinate data were originally acquired from the Blue Brain Project Cell Atlas.Usage: Extract this archive and place the microdata folder directly into the Functional consistency/ directory of the NIGC GitHub repository. It is required to run the process_neuron_data.py preprocessing pipeline.2. NIGC_matrices.zip
Contents: This archive contains two key components from our structural similarity validation experiments:connection_progress.npy: A large intermediate cache file storing the generated synaptic connectivity states for the mouse primary visual cortex (V1) microcircuit.neuron_count_experiment_results/: A directory containing the pre-computed connection matrices and topological metrics for the multi-scale neuron count benchmarking experiments (detailed in the Supplementary Information).Usage: These files cache the results of the most computationally expensive synaptic querying and generation processes. Extract this archive and place connection_progress.npy into the Structural similarity/visual_cortex_result/ directory, and the neuron_count_experiment_results folder into the Structural similarity/ directory of the repository. Related Code The official implementation code utilizing this dataset can be found on GitHub: https://github.com/jiedia/NIGC