Simulation-based inference of cell migration dynamics in complex spatial environments.

Area of Science:

• Systems biology
• Cell biology
• Biophysics

Background:

• Microfabricated chips and advanced imaging are used to study cell migration in constrained environments.
• Computational models are crucial for understanding how physical geometry impacts intracellular dynamics.

Purpose of the Study:

• To integrate experimental data of dendritic cell migration in constrained microenvironments with a Cellular Potts model.
• To develop advanced inference techniques for decoding complex cell migration behaviors.

Main Methods:

• Utilized microfabricated chips to create geometrically constrained environments for dendritic cell migration.
• Integrated experimental observations into a Cellular Potts model.
• Applied neural posterior estimation with in-the-loop learning for parameter inference.

Main Results:

• Spatial constraints were shown to modulate cell motility dynamics, including speed and directional changes.
• Classical statistics (e.g., mean squared displacement) were insufficient for capturing rich spatiotemporal patterns.
• The developed neural posterior estimation method enabled robust and flexible parameter inference.

Conclusions:

• The study provides a data-driven framework for calibrating computational models of cell migration.
• The findings advance quantitative analysis of cell migration in structured microenvironments.
• The novel inference approach enhances the understanding of cell behavior in complex spatial settings.