Skip to content

README_en.md

Latest commit

 

History

History
134 lines (99 loc) · 3.78 KB

README_en.md

File metadata and controls

134 lines (99 loc) · 3.78 KB
Raw

Spherical Brain Space Simulator

English Version | 中文版

Original Declaration: This concept was first proposed and publicly released by rubyguyan in March 2026.

For detailed concept description, see: IDEA.md

A brain space simulator based on FlyWire/FlyConnectome real neuron data.

Introduction

This project uses real neuron morphology data (SWC format) from the fruit fly brain to simulate neural networks in a spherical 3D space. Neurons automatically form synaptic connections through collisions, simulating real neural network activity.

Core Innovations

  • Spherical Space Constraint: Neurons are constrained within a spherical space, forming a natural spatial gradient
  • Real Neuron Data: Uses real neuron morphology from the FlyWire project
  • Collision-Driven Connections: Neurons automatically form synapses through physical collisions
  • Nutrient Supply System: Nutrients are abundant at the center, scarce at the edges
  • Neuron Lifecycle: Division, death, and dynamic balance

Data Sources

  • FlyWire - Fruit fly whole-brain connectome project
  • FAFB (Full Adult Fly Brain) - Adult fruit fly whole-brain dataset
  • Neuron IDs from the FlyWire database

Features

  • Real SWC neuron file loading
  • LIF (Leaky Integrate-and-Fire) neuron model
  • Neuron signal sending/receiving interface
  • Automatic synaptic connection through collision
  • Nutrient supply system (spatial gradient)
  • 2D/3D visualization switching
  • Dynamic space size adjustment
  • Time scale speed control
  • Automatic cluster identification
  • Signal monitoring panel

Installation

pip install matplotlib numpy

Usage

python brain_space.py

Command Line Options

python brain_space.py --count 10        # Number of neurons
python brain_space.py --category name   # Specify neuron category
python brain_space.py --radius 1000     # Spherical space radius

Keyboard Shortcuts

Key Function
Space Pause/Resume
2/3 Switch 2D/3D mode
[ ] Shrink/Expand space
- = Slow down/Speed up
H Show help window
I Show/Hide signal panel
N Add new neuron
E Send input to neuron
S Save state
T Stimulate random neuron
Q Quit

File Structure

spherical-brain-space/
├── brain_space.py      # Main program
├── swc_neuron.py       # SWC neuron module
├── IDEA.md             # Concept description
├── README.md           # Chinese documentation
├── README_en.md        # English documentation
├── neurons/            # Neuron SWC files
│   └── *.swc          # Real neuron morphology data
├── docs/               # Documentation
└── brain_space_data/   # Runtime data

SWC File Format

Standard SWC neuron morphology format:

  • Each line: Point ID, Label, X, Y, Z, Radius, Parent
  • Labels: 0=undefined, 1=soma, 5=fork point, 6=end point

Technical Details

Coordinate Transformation

FAFB coordinates (nanometers) to spherical space coordinates:

  • X: 350000-550000 nm
  • Y: 80000-220000 nm
  • Z: 20000-240000 nm

LIF Neuron Model Parameters

  • Resting potential: -70 mV
  • Threshold potential: -55 mV
  • Reset potential: -75 mV
  • Refractory period: 2 ms

Citation

If you use this concept or code in your research or project, please cite:

Spherical Brain Space Simulator
Author: rubyguyan
GitHub: https://github.com/rubyguyan/spherical-brain-space
Release Date: March 2026

License

  • Code: MIT License - Free to use, please credit the source
  • Concept Description: CC BY 4.0 - Please credit the source when reposting

Acknowledgments

  • FlyWire - Fruit fly whole-brain connectome project
  • navis - Neuron morphology analysis tool