Sparse polynomial surrogates for F-actin networks with compliant crosslinkers

Area of Science:

• Biophysics
• Computational Biology
• Materials Science

Background:

• Filamentous actin (F-actin) networks are crucial for cell elasticity and structure.
• Current mechanical models struggle with biological variability and dynamic processes.
• Understanding F-actin mechanics is vital for cell motility, division, and disease.

Purpose of the Study:

• Develop a stochastic modeling framework for F-actin networks.
• Integrate uncertainty quantification and sensitivity analysis into F-actin models.
• Enable accurate predictions of F-actin network behavior under varying conditions.

Main Methods:

• Utilized Polynomial Chaos Expansion (PCE) surrogates with Finite Element Method (FEM).
• Replaced filament-scale equations with PCE surrogates for efficient analysis.
• Incorporated statistical moments into a micro-sphere network model.

Main Results:

• The PCE surrogate model accurately predicted stress quantities with <1% error compared to Monte Carlo simulations.
• The model successfully captured network variability (second-order moments).
• Demonstrated rapid, statistically faithful predictions for various deformation and rheology scenarios.

Conclusions:

• The proposed framework offers a scalable method for F-actin mechanical modeling.
• It effectively incorporates intrinsic material variability into computational models.
• Has significant implications for studying cell dynamics and cytoskeletal remodeling-related pathologies.