Welcome to MacTrack2’s documentation!

Derived from the work by Axel de Montgolfier during his time at the LPHI lab at the University of Montpellier. Our work aims to simplify its use by adding various scripts. For instance, a gettingstarted file helps with the analysis of your videos, based on examples we provide in the examples folder. You can also create your own model by running the quickstart.py file.

Make sure to carefully read the documentation for both files. You can find it either in the source files or on the website available here.

The goal of this project is to accurately segment macrophages. This segmentation enables superimposing results on the green channel video to analyze calcium flashes of the tracked macrophages. This way, we can retrieve information about each macrophage — such as flash amplitude and intensity — without being hindered by the noise commonly present in microscopic images.

Installation

As my internship followed Axel’s work, I had to familiarize myself with his original project. This repository contains upgrades to the original work, available here. We recommend cloning this repository as it includes various improvements.

Clone the repository:

git clone https://github.com/guibouland/MacTrack2.git

You can then install the dependencies listed in requirements.txt. We recommend creating a virtual environment.

If you use (micro)mamba:

micromamba create -n mactrack-env python=3.12   # Create virtual env using Python 3.12
micromamba activate mactrack-env                # Activate the environment
cd MacTrack2                                    # Enter the project directory
pip install -r requirements.txt                 # Install dependencies

If you use Python’s venv module:

python3.12 -m venv mactrack-env                  # Create virtual environment
source mactrack-env/bin/activate                 # Activate on Linux/macOS
mactrack-env\Scripts\activate                    # Activate on Windows
cd MacTrack2
pip install -r requirements.txt                  # Install dependencies

You’re now ready to go!

To try the module we provide, run the gettingstarted file. It uses a pre-trained model found in the examples folder. If you want to train your own model, use quickstart.py.

Create Your Own Dataset and Model

For more details on how to create your own dataset, train and test a model, and apply it to videos beyond those provided, refer to the documentation available here.

Materials and Methods

All videos and frames used for analysis and segmentation were acquired using a spinning disk confocal microscope. It illuminates macrophages in red and calcium flashes in green, using a zebrafish transgenic lineage.

We amputated the caudal fin fold of a zebrafish to study potential correlations between macrophage polarization and calcium flashes.

Contributors

Acknowledgement

This project relies mainly on the kartezio Python package, developed by Kévin Cortacero.

Contents: