TrajPy: empowering feature engineering for trajectory analysis across domains

TL;DR

TrajPy is a user-friendly, open-source Python package designed to facilitate trajectory analysis across various domains by providing physical descriptors and a graphical interface, enhancing the analysis of complex path data.

Contribution

This work introduces TrajPy, the first comprehensive, accessible Python toolkit for trajectory analysis with physical descriptors and visualization, applicable across multiple scientific fields.

Findings

Successfully applied to mitochondrial motility studies in neuroblastoma cells

Used for analyzing in silico models of cell migration

Provides an easy-to-use GUI and physical descriptors for trajectory characterization

Abstract

Trajectories, sequentially measured quantities that form a path, are an important presence in many different fields, from hadronic beams in physics to electrocardiograms in medicine. Trajectory anal-ysis requires the quantification and classification of curves either using statistical descriptors or physics-based features. To date, there is no extensive and user-friendly package for trajectory anal-ysis available, despite its importance and potential application across domains. We developed a free open-source python package named TrajPy as a complementary tool to empower trajectory analysis. The package showcases a friendly graphic user interface and provides a set of physical descriptors that help characterizing these intricate structures. In combina-tion with image analysis, it was already successfully applied to the study of mitochondrial motility in neuroblastoma cell lines and to the analysis of in silico models for cell migration. The TrajPy package was developed in Python 3 and released under the GNU GPL-3 license. Easy installation is available through PyPi and the development source code can be found in the repository https://github.com/ocbe-uio/TrajPy/. The package release is automatically archived under the DOI 10.5281/zenodo.3656044.