Sign in Join

Live chat for Free Pascal, Lazarus, Delphi and every Pascal programmer.

Catalog · How the catalog works

Random project

NETrex

Neural Network Modeling and Dynamic Clamp Software

View on GitHub

Anyone may press Interesting, signed in or not. It helps others find this project.

Member rating

No ratings yet.

Sign in to rate this project.

README

NETrex — Dynamic Clamp and Network Simulator

NETrex is a real-time neural network modeling and dynamic clamp application for neuroscience research. It combines the functionality of the legacy NetSim (simulation) and NetClamp (dynamic clamp) applications into a single unified program.

Key features:

  • Build networks of model neurons with Hodgkin-Huxley conductances, synapses, gap junctions, and current injection electrodes
  • Run pure simulations or real-time dynamic clamp experiments with NI-DAQmx hardware
  • Real-time trace plotting, CSV data export, and configurable timing protocols
  • Cross-platform: works on Windows, Linux, and macOS

Author: E. Brady Trexler, Ph.D.
License: GNU General Public License v3
Website: hudsonvalleysci.com

Download & Install

Pre-built binaries are available on the Releases page.

Windows (recommended for dynamic clamp experiments)

  1. Download NETrex-Windows-x64.zip from the latest release
  2. Extract the zip to a folder (e.g., C:\NETrex)
  3. Double-click NETrex.exe to run
  4. For dynamic clamp with NI hardware: Install the free NI-DAQmx Runtime from ni.com. NETrex will automatically detect your DAQ device.
  5. For simulation only: No additional software needed — just run the app

macOS

  1. Download NETrex-macOS.zip from the latest release

  2. Extract the zip — you'll get NETrex.app

  3. Drag NETrex.app to your Applications folder

  4. Before first launch, open Terminal and run:

    bash
    xattr -cr /Applications/NETrex.app

    This removes the macOS quarantine flag that blocks unsigned apps downloaded from the internet.

  5. Double-click NETrex.app to run

  6. Note: macOS runs in simulation mode only (no NI-DAQmx hardware support). A mock DAQ backend is included for testing dynamic clamp networks.

Linux

  1. Download NETrex-Linux-x64.zip from the latest release
  2. Extract and make executable: chmod +x NETrex
  3. Run: ./NETrex
  4. Requires Qt 5 libraries (sudo apt install libqt5widgets5 on Ubuntu/Debian)
  5. For dynamic clamp: Install the NI-DAQmx driver for Linux

Quick Start (build from source)

bash
git clone <repo-url> NETrex
cd NETrex
mkdir build && cd build
cmake ..
make NETrex
./NETrex

Build Requirements

Dependency Required Notes
C++17 compiler Yes GCC 7+, Clang 5+, MSVC 2017+
CMake 3.15+ Yes Build system
Qt 5 (Widgets) Yes GUI framework
NI-DAQmx driver No Only needed for real hardware dynamic clamp

All other dependencies (nlohmann/json, NI-DAQmx headers) are included in the repository.

Platform-Specific Instructions

Windows (with NI-DAQmx hardware)

This is the primary target for dynamic clamp experiments.

  1. Install prerequisites:

  2. Set up environment:

    cmd
    set CMAKE_PREFIX_PATH=C:\Qt\5.15.2\msvc2019_64

    Adjust the path to match your Qt installation.

  3. Build:

    cmd
    git clone <repo-url> NETrex
cd NETrex
mkdir build && cd build
cmake .. -G "Visual Studio 17 2022" -A x64
cmake --build . --config Release --target NETrex

  4. Run:

    cmd
    Release\NETrex.exe

    CMake will automatically detect the NI-DAQmx library and enable hardware support. You should see this during cmake configuration:

    pascal
    -- NI-DAQmx found: C:/.../nidaqmx.lib

  5. Deploy (create standalone folder):

    cmd
    mkdir deploy
copy Release\NETrex.exe deploy\
cd deploy
windeployqt NETrex.exe

    The deploy folder can be zipped and shared — it contains all needed Qt DLLs.

Linux (with NI-DAQmx hardware)

NI-DAQmx is supported on select Linux distributions (RHEL, CentOS, Ubuntu, openSUSE).

  1. Install prerequisites:

    bash
    # Ubuntu/Debian
sudo apt install build-essential cmake qt5-default

# Fedora/RHEL
sudo dnf install gcc-c++ cmake qt5-qtbase-devel

  2. Install NI-DAQmx (optional, for hardware): Download the Linux driver from ni.com. Follow NI's installation instructions. The library installs to /usr/lib/x86_64-linux-gnu/libnidaqmx.so.

  3. Build:

    bash
    git clone <repo-url> NETrex
cd NETrex
mkdir build && cd build
cmake ..
make NETrex

    If NI-DAQmx is installed, cmake will report:

    pascal
    -- NI-DAQmx found: /usr/lib/x86_64-linux-gnu/libnidaqmx.so

    If not installed, it will use the mock backend for testing.

  4. Run:

    bash
    ./NETrex

macOS (development/simulation only)

NI-DAQmx is not available on macOS. NETrex runs with a mock DAQ backend that simulates a biological cell with Hodgkin-Huxley currents, allowing full development and testing of networks.

  1. Install prerequisites:
    bash
    # Install Xcode command line tools
xcode-select --install

# Install Homebrew packages
brew install cmake qt@5
  2. Build:
    bash
    git clone <repo-url> NETrex
cd NETrex
mkdir build && cd build
cmake .. -DCMAKE_PREFIX_PATH=$(brew --prefix qt@5)
make NETrex
  3. Run:
    bash
    ./NETrex

How It Works

Simulation Mode

When a network contains only model cells (no biological cells), NETrex runs as a pure simulator. The network is updated at the requested sample rate using RK4 integration, and results are plotted in real time.

Dynamic Clamp Mode

When a network contains one or more biological cells, NETrex automatically switches to dynamic clamp mode:

  1. Analog input channels read membrane voltages from the amplifier
  2. The network model computes synaptic/intrinsic currents based on the measured voltages
  3. Computed currents are written to analog output channels and injected into the cell

The timing protocol supports three phases:

  • Before: DAQ is reading but no current is injected (baseline)
  • During: Active dynamic clamp — currents are computed and injected
  • After: DAQ is reading but current injection stops (recovery)

Supported Cell Types

  • Model Cell — Integrate-and-fire neuron with configurable currents
  • Vm Playback Cell — Plays back a voltage waveform from a file
  • Biological Cell — Represents a real neuron connected via DAQ hardware

Supported Current/Synapse Types

  • Hodgkin-Huxley conductance (HH, HH2)
  • Gap junction
  • Voltage clamp PID
  • Generic bidirectional synapse
  • Playback current waveform

File Format

Networks are saved as JSON files (.json). Example:

json
{
  "cells": [
    {
      "name": "Cell 1",
      "type": "Model Cell",
      "active": true,
      "currents": [...]
    }
  ],
  "electrodes": [...],
  "synapses": [...]
}

Troubleshooting

cmake can't find Qt5: Set CMAKE_PREFIX_PATH to your Qt installation:

bash
cmake .. -DCMAKE_PREFIX_PATH=/path/to/qt5

NI-DAQmx not detected: Ensure the full NI-DAQmx driver is installed (not just Runtime). Check that libnidaqmx is in your library path. cmake will print a status message indicating whether the library was found.

Mock backend on Windows/Linux: If you want to develop without hardware, NETrex works fine — it just uses the mock backend. No configuration needed.

Building Redistributable Binaries

Windows

cmd
cmake --build . --config Release --target NETrex
mkdir deploy
copy Release\NETrex.exe deploy\
cd deploy
windeployqt NETrex.exe

Zip the deploy folder. Recipients need the NI-DAQmx Runtime (free download from ni.com) if using hardware.

Linux

bash
# Build with static Qt or use linuxdeployqt:
# https://github.com/probonopd/linuxdeployqt
make NETrex
linuxdeployqt NETrex -appimage

Produces a portable AppImage. Recipients need NI-DAQmx driver for hardware.

macOS

bash
make NETrex
macdeployqt NETrex.app -dmg

Produces a .dmg for distribution (simulation only, no DAQ hardware support).