LapTrack

Features

Provides a robust particle tracking algorithm using the Linear Assignment Problem, with various cost functions for linking.

See the publication and associated repository for the algorithm and parameter optimization by Ray-Tune.

Requirements

Python >= 3.8 is supported. The software is tested against Python 3.8-3.12 in Ubuntu, and 3.12 in MacOS and Windows environments, but the other combinations should also be fine. Please file an issue if you encounter any problem.

Installation

You can install LapTrack via pip from PyPI:

$ pip install laptrack

In Google Colaboratory, try

$ pip install --upgrade laptrack spacy flask matplotlib

to update the pre-installed packages.

Usage

Please see the Usage for details. The example notebooks are provided in docs/examples.

notebook name	short description	Google Colaboratory
api_example.ipynb	Introducing the package API by a simple example.
bright_spots.ipynb	Application example: detecting bright spots by scikit-image blob_log and tracking them.
cell_segmentation.ipynb	Application example: tracking centroids of the segmented C2C12 cells undergoing divisions.
napari_interactive_fix.ipynb	Illustrates the usage of the ground-truth-preserved tracking with napari.
overlap_tracking.ipynb	Illustrates the usage of the custom metric to use segmentation overlaps for tracking.

The API reference covers the main classes and functions provided by LapTrack.

Contributing

Contributions are very welcome. To learn more, see the Contributor Guide.

License

Distributed under the terms of the The 3-Clause BSD License, LapTrack is free and open source software.

Issues

If you encounter any problems, please file an issue along with a detailed description.

Credits

This program implements a modified version of the algorithm in the K. Jaqaman et al. (2008).
Inspired by TrackMate a lot. See documentation for its detailed algorithm, the 2016 paper, and the 2021 paper.
The data in docs/examples/napari_interactive_fix_data are generated by cropping images in 10.5281/zenodo.6087728, which is distributed with Creative Commons Attribution 4.0 International.
The data in docs/examples/cell_segmentation_data are generated by cropping and resizing images in https://osf.io/ysaq2/, which is distributed with Creative Commons Attribution 4.0 International. See 10.1038/sdata.2018.237 for details.
The data in docs/examples/overlap_tracking_data is generated by cropping segmentation.npy in https://github.com/NoneqPhysLivingMatterLab/cell_interaction_gnn, which is distributed with Apache License 2.0. See the original paper for details.
The data in docs/examples/3D_tracking_data is generated by resizing iamges in https://bbbc.broadinstitute.org/BBBC050 , which is distributed with Creative Commons Attribution 3.0 Unported License. See 10.1038/s41540-020-00152-8 for details.
This project was generated from @cjolowicz’s Hypermodern Python Cookiecutter template.

Citation

If you use this program for your research, please cite it and help us build more.

@article{fukai_2022,
  title = {{{LapTrack}}: Linear Assignment Particle Tracking with Tunable Metrics},
  shorttitle = {{{LapTrack}}},
  author = {Fukai, Yohsuke T and Kawaguchi, Kyogo},
  year = {2022},
  month = dec,
  journal = {Bioinformatics},
  pages = {btac799},
  issn = {1367-4803},
  doi = {10.1093/bioinformatics/btac799},
}

@misc{laptrack,
   author = {Yohsuke T. Fukai},
   title = {laptrack},
   year  = {2021},
   url   = {https://doi.org/10.5281/zenodo.5519537},
}