ShapeSpaceExplorer: Analysis of morphological transitions in migrating cells using similarity-based shape space...

Area of Science:

• Computational biology
• Biophysics
• Image analysis

Background:

• Cell migration is crucial for physiological and pathological processes.
• Cell shape dynamics are emergent properties of the forces driving cell migration.
• Understanding cell shape-migration relationships is key to studying cellular behavior.

Purpose of the Study:

• To introduce ShapeSpaceExplorer, an interactive software for analyzing complex 2D shape series.
• To demonstrate the software's application in analyzing cell morphology changes during cell migration.
• To develop a machine learning approach for understanding cell shape dynamics and migration behavior.

Main Methods:

• Developed ShapeSpaceExplorer software for interactive extraction, visualization, and analysis of 2D shape series.
• Implemented a machine learning algorithm to analyze cell shape from time-lapse images and learn the intrinsic low-dimensional structure of cell shape space.
• Introduced a novel, rapid, and landmark-free shape difference measure for unbiased analysis of diverse cell morphologies.

Main Results:

• The software enables visualization of cell shape distribution differences after perturbation experiments.
• Quantitative relationships between cell shape and migration behavior were analyzed.
• The method successfully predicted cell turning based on dynamic cell shape information.

Conclusions:

• ShapeSpaceExplorer provides a powerful tool for visualizing and analyzing cell morphology changes.
• The machine learning approach effectively maps cell shape dynamics and predicts migration behavior.
• The software has broad applicability to various biological and inanimate object shape analyses.