Skip to content

Getting Started

Requires Python 3.12+

Check your version with python --version.

System requirements

NEURON and an MPI library must be available before installing MyoGen.

Platform Before installing MyoGen
Windows NEURON 8.2.7 — download, run the installer, select "Add to PATH".
Linux sudo apt install libopenmpi-dev (Ubuntu/Debian) or sudo dnf install openmpi-devel (Fedora).
macOS brew install open-mpi.

Windows prerequisites

You must install the following before installing MyoGen on Windows:

  1. Visual C++ Build Tools — install Visual C++ Build Tools with: MSVC Build Tools for x64/x86 (latest), MSVC v143 (VS 2022 C++ x64/x86), Windows 11 SDK (latest), and C++ core desktop features.
  2. NEURON 8.2.7 — run the installer, select "Add to PATH", then restart your terminal.

Install

We recommend uv, a fast Python package manager.

# install uv (Linux/macOS)
curl -LsSf https://astral.sh/uv/install.sh | sh
# ...or Windows PowerShell:
# powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

# in a fresh project
uv init my_emg_project && cd my_emg_project
uv add myogen

pip install myogen

git clone https://github.com/NsquaredLab/MyoGen.git
cd MyoGen
uv sync
uv run poe setup_myogen

Verify the install:

uv run python -c "from myogen import simulator; print('MyoGen installed successfully!')"

Optional: GPU acceleration

For 5–10× faster convolutions on an NVIDIA GPU: uv add cupy-cuda12x.

Quick start

Generate motor-unit action potentials and surface EMG:

from myogen import simulator
import quantities as pq

# 1. Recruitment thresholds for 100 motor units
thresholds, _ = simulator.RecruitmentThresholds(
    N=100, recruitment_range__ratio=50, mode="fuglevand"
)

# 2. Muscle model with fiber distribution
muscle = simulator.Muscle(
    recruitment_thresholds=thresholds,
    radius_bone__mm=1.0 * pq.mm,
    fiber_density__fibers_per_mm2=400 * pq.mm**-2,
    fat_thickness__mm=10 * pq.mm,
    autorun=True,
)

# 3. Surface electrode array
electrode_array = simulator.SurfaceElectrodeArray(
    num_rows=5,
    num_cols=5,
    inter_electrode_distance__mm=5 * pq.mm,
    electrode_radius__mm=5 * pq.mm,
    bending_radius__mm=muscle.radius__mm + muscle.skin_thickness__mm + muscle.fat_thickness__mm,
)

# 4. Surface EMG simulator
surface_emg = simulator.SurfaceEMG(
    muscle_model=muscle,
    electrode_arrays=[electrode_array],
    sampling_frequency__Hz=2048.0,
    MUs_to_simulate=[0, 1, 2, 3, 4],
)

# 5. Simulate MUAPs and access the results
muaps = surface_emg.simulate_muaps()          # neo.Block of MUAP templates
first_mu = muaps.segments[0].analogsignals[0]  # everything is inspectable

For the full pipeline — spike trains, force, intramuscular EMG, spinal networks — browse the Examples gallery.