ultrack 0.8.0

Large-scale cell tracking under segmentation uncertainty.

Overview

Ultrack is a versatile and scalable cell tracking method designed to address the challenges of tracking cells across 2D, 3D, and multichannel timelapse recordings, especially in complex and crowded tissues where segmentation is often ambiguous. By evaluating multiple candidate segmentations and employing temporal consistency, Ultrack ensures robust performance under segmentation uncertainty. Ultrack's methodology is explained here.

https://github.com/royerlab/ultrack/assets/21022743/10aace9c-0e0e-4310-a103-f846683cfc77

Zebrafish imaged using DaXi whole embryo tracking.

Features

• Versatile Cell Tracking: Supports 2D, 3D, and multichannel datasets.
• Robust Under Segmentation Uncertainty: Evaluates multiple candidate segmentations.
• High Performance: Scales from small in vitro datasets to terabyte-scale developmental time-lapses.
• Integration: Compatible with FiJi, napari, and high-performance clusters via SLURM.

Installation

pip install ultrack

We recommend installing into a virtual environment to avoid package conflicts. gurobi is optional but recommended for best performance; see the installation guide for details on Gurobi setup and GPU acceleration.

Optionally, we provide multiple Docker images. For instructions, see the docker folder.

Usage

Here is a basic example to get you started:

import napari
from ultrack import MainConfig, Tracker

# __main__ is recommended to avoid multi-processing errors
if __name__ == "__main__":
      # Load your data
      foreground = ...
      contours = ...

      # Create config
      config = MainConfig()

      # Run tracking
      tracker = Tracker(config)
      tracker.track(foreground=foreground, edges=contours)

      # Visualize results in napari
      tracks, graph = tracker.to_tracks_layer()
      napari.view_tracks(tracks[["track_id", "t", "z", "y", "x"]], graph=graph)
      napari.run()

More usage examples can be found here, including their environment files and installation instructions.

Documentation

Comprehensive documentation is available here.

These additional developer documentation are available:

• Parameter configuration schema.
• Intermediate segmentation and tracking SQL database are here.

Gurobi Setup

Install Gurobi using Conda

In your existing Conda environment, install Gurobi with the following command:

conda install -c gurobi gurobi

Obtain and Activate an Academic License

1. Register at Gurobi's website with your academic email.
2. Navigate to the Gurobi's named academic license page
3. Follow the instructions to get your license key.
4. Activate your license, In your Conda environment, run:

grbgetkey YOUR_LICENSE_KEY

5. Replace YOUR_LICENSE_KEY with the key you received. Follow the prompts to complete activation.

Verify Installation

Verify Gurobi's installation by running:

ultrack check_gurobi

Depending on the operating system, the gurobi library might be missing and you need to install it from here.

Who is using Ultrack?

You can find a list of projects and papers that have used ultrack on this page.

Contributing

We welcome contributions from the community! To get started, please read our contributing guidelines. Then, report issues and submit pull requests on GitHub.

License

This project is licensed under the BSD-3 License - see the LICENSE file for details.

Citing

If you use ultrack in your research, please cite the following papers, the algorithm and the biological applications and software.

@inproceedings{bragantini2024ucmtracking,
  title={Large-scale multi-hypotheses cell tracking using ultrametric contours maps},
  author={Bragantini, Jord{\~a}o and Lange, Merlin and Royer, Lo{\"\i}c},
  booktitle={European Conference on Computer Vision},
  pages={36--54},
  year={2024},
  organization={Springer}
}

@article{bragantini2024ultrack,
  title={Ultrack: pushing the limits of cell tracking across biological scales},
  author={Bragantini, Jord{~a}o and Theodoro, Ilan and Zhao, Xiang and Huijben, Teun APM and Hirata-Miyasaki, Eduardo and VijayKumar, Shruthi and Balasubramanian, Akilandeswari and Lao, Tiger and Agrawal, Richa and Xiao, Sheng and others},
  journal={bioRxiv},
  pages={2024--09},
  year={2024},
  publisher={Cold Spring Harbor Laboratory}
}

And the respective auxiliary methods according to your use, a non-exhaustive list being:

• napari
• cellpose
• stardist
• fiji

Acknowledgements

We acknowledge the contributions of the community and specific individuals. Detailed acknowledgments can be found in our documentation.