Learning dynamical models of single and collective cell migration: a review

Area of Science:

• Physics, Biophysics, Computational Biology

Background:

• Cell migration is crucial for development, immunity, and disease.
• Understanding cell migration requires physical models that capture complex, stochastic behaviors.
• Connecting theoretical models to experimental data remains a key challenge.

Purpose of the Study:

• To review recent advances in data-driven theoretical approaches for modeling stochastic cell migration.
• To discuss the integration of large experimental datasets with physical models across scales.
• To highlight how data-driven methods reveal molecular mechanisms underlying cell behavior.

Main Methods:

• Reviewing data-driven inference techniques for dynamical models of cell migration.
• Analyzing approaches for inferring underdamped stochastic equations of motion from experimental data.
• Examining applications of machine learning to cell heterogeneity and collective dynamics.

Main Results:

• Data-driven methods successfully infer dynamical models of stochastic cell migration.
• These approaches link experimental observations to physical models from single-cell to tissue scales.
• Machine learning and data inference reveal insights into cell heterogeneity and collective behaviors.

Conclusions:

• Data-driven approaches are vital for building accurate physical models of cell migration.
• These methods provide conceptual links between molecular mechanisms and emergent cell behaviors.
• This interdisciplinary approach advances our understanding of fundamental biological processes.