trajpy

Trajectory classifier for cells, nanoparticles & whatelse.


Keywords
trajectory, quantification, feature, engineering, diffusion, classification, cell-migration, python-package, trajectory-analysis
License
GPL-3.0
Install
pip install trajpy==1.3.1

Documentation

PyPI version Maintainability codecov Build Status PyUp Python 3 Documentation Status License: GPL v3 DOI

TrajPy

Trajectory classification is a challenging task and fundamental for analysing the movement of nanoparticles, bacteria, cells and active matter in general.

We propose TrajPy as an easy pythonic solution to be applied in studies that demand trajectory classification. It requires little knowledge of programming and physics to be used by nonspecialists.

TrajPy is composed of three main units of code:

  • The training data set is built using a trajectory generator
  • Features are computed for characterizing the trajectories
  • The classifier built on Scikit-Learn.

Our dataset and Machine Learning (ML) model are available for use, as well the generator for building your own database.

Installation

We have the package hosted at PyPi, for installing use the command line:

pip3 install trajpy

If you want to test the development version, clone the repository at your local directory from your terminal:

git clone https://github.com/phydev/trajpy

Then run the setup.py for installing

python setup.py --install

Basic Usage Example

First we import the package

import trajpy.trajpy as tj

Then we load the data sample provided in this repository, we pass the arguments skip_header=1 to skip the first line of the file and delimiter=',' to specify the file format

filename = 'data/samples/sample.csv'
r = tj.Trajectory(filename,
                  skip_header=1,
                  delimiter=',')

Finally, for computing a set of features for trajectory analysis we can simple run the function r.compute_features()

    r.compute_features()

The features will be stored in the object r, for instance:

  >>> r.asymmetry
  >>> 0.5782095322093505
  >>> r.fractal_dimension
  >>> 1.04
  >>> r.efficiency
  >>> 0.29363293632936327
  >>> r.gyration_radius
  >>> array([[30.40512689,  5.82735002,  0.96782673],
  >>>     [ 5.82735002,  2.18625318,  0.27296851],
  >>>     [ 0.96782673,  0.27296851,  2.41663589]])

For more examples please consult the extended documentation: https://trajpy.readthedocs.io/

Requirements

  • numpy >= 1.14.3
  • scipy >= 1.5.4

[ ~ Dependencies scanned by PyUp.io ~ ]

Citation

If using the TrajPy package in academic work, please cite Moreira-Soares et al. (2020), in addition to the relevant methodological papers.

@article{moreira2020adhesion,
  title={Adhesion modulates cell morphology and migration within dense fibrous networks},
  author={Moreira-Soares, Maur{\'\i}cio and Cunha, Susana P and Bordin, Jos{\'e} Rafael and Travasso, Rui DM},
  journal={Journal of Physics: Condensed Matter},
  volume={32},
  number={31},
  pages={314001},
  year={2020},
  publisher={IOP Publishing}
}

@software{mauricio_moreira_2020_3978699,
  author       = {Mauricio Moreira and Eduardo Mossmann},
  title        = {phydev/trajpy: TrajPy 1.3.1},
  month        = aug,
  year         = 2020,
  publisher    = {Zenodo},
  version      = {1.3.1},
  doi          = {10.5281/zenodo.3978699},
  url          = {https://doi.org/10.5281/zenodo.3978699}
}

Contribution

This is an open source project, and all contributions are welcome. Feel free to open an Issue, a Pull Request, or to e-mail us.

Publications using trajpy

Simões, RF, Pino, R, Moreira-Soares, M, et al. Quantitative Analysis of Neuronal Mitochondrial Movement Reveals Patterns Resulting from Neurotoxicity of Rotenone and 6-Hydroxydopamine. FASEB J. 2021; 35:e22024. doi:10.1096/fj.202100899R

Moreira-Soares, M., Pinto-Cunha, S., Bordin, J. R., Travasso, R. D. M. Adhesion modulates cell morphology and migration within dense fibrous networks. https://doi.org/10.1088/1361-648X/ab7c17

References

Arkin, H. and Janke, W. 2013. Gyration tensor based analysis of the shapes of polymer chains in an attractive spherical cage. J Chem Phys 138, 054904.

Wagner, T., Kroll, A., Haramagatti, C.R., Lipinski, H.G. and Wiemann, M. 2017. Classification and Segmentation of Nanoparticle Diffusion Trajectories in Cellular Micro Environments. PLoS One 12, e0170165.