Installation

Installation

There are two ways to install the SONLab FRET Analysis Tool:

1. Using the installers (recommended) — automated setup with a desktop launcher.
2. Manual installation — for advanced users who want full control of the environment.

The desktop launcher created by the installer.

---

1. Installers (recommended)

The installers/ directory in the repository contains a script for each platform that creates a virtual environment, installs dependencies, and adds a desktop launcher.

Platform	Script
Windows	installers/install_windows.ps1
Linux	installers/install_linux.sh
macOS	installers/install_mac.sh

See installers/README.md for step-by-step instructions specific to each platform.

---

2. Manual installation

Prerequisites

All platforms

• Python 3.10 — required for dependency compatibility. Download Python 3.10.11 and enable Add Python to PATH during installation. Newer Python versions are not supported because of dependency constraints.
• pip (Python package manager)
• Git (or download the repository as a ZIP)
• At least 8 GB free disk space and an internet connection

Linux (additional): build tools and Python development headers. macOS (additional): Xcode Command Line Tools; Homebrew is recommended for installing Python.

Steps

1. Get the source

git clone https://github.com/sonlab-metu/SONLab-FRET-Tool.git
cd SONLab-FRET-Tool

2. Create and activate a virtual environment

Windows (Command Prompt):

python -m venv venv
.\venv\Scripts\activate

Linux/macOS:

python3 -m venv venv
source venv/bin/activate

3. Install the core dependencies

pip install -r installers/requirements.txt

4. Install PyTorch for your hardware

Choose the command that matches your compute platform:

Hardware	Command
NVIDIA (CUDA 11.8)	pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
NVIDIA (CUDA 12.6)	pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126
NVIDIA (CUDA 12.8)	pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
AMD ROCm 6.3 (Linux)	pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm6.3
CPU only	pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu

On Apple Silicon, use the standard CPU command; PyTorch automatically uses the Metal Performance Shaders (MPS) backend.

5. Run the application

python3 -m GUI.main_gui

Note: the manual method does not create a desktop shortcut. Activate the virtual environment and run the command each time.

---

System recommendations

• Windows 10/11, macOS 10.15+, or a recent Linux distribution.
• Minimum 8 GB RAM (16 GB recommended for large datasets or batch processing).
• A CUDA-capable GPU significantly speeds up Cellpose segmentation but is not required.
• Screen resolution of 1920×1080 or higher is recommended for the multi-panel layout.

If you run into problems, see Troubleshooting and FAQ.