A fictitious domain method with a hybrid cell model for simulating motion of cells in fluid flow

Guang Lin, Wenrui Hao, Zhiliang Xu, Chun Liu, Guang Lin 0001Journal of Computational Physics 2015 · DOI

TL;DR

This paper introduces a novel hybrid cell model and a fictitious domain method to simulate the complex interactions between biological cells and fluid flow. This approach accurately captures cell mechanics and fluid dynamics, offering a powerful tool for understanding various biological processes.

What Problem Does This Paper Address?

Understanding the dynamics of biological cells in fluid flow is crucial for many biological and bio-mechanical processes, such as blood cell interactions, tumor metastasis, and bacterial motility. These phenomena involve intricate fluid-structure interactions where the cell's membrane and internal structure play a vital role. Accurately simulating these interactions requires robust numerical methods that can handle both the fluid dynamics (governed by Navier-Stokes equations) and the complex mechanics of the cell membrane. Existing methods often struggle to simultaneously model the cell's internal structure, its deformable membrane, and its interaction with the surrounding fluid in a computationally efficient and accurate manner. This paper aims to bridge this gap by developing a hybrid cell model and a fictitious domain approach that can handle these complexities.

Methods

The core of this work lies in two main components: a hybrid cell model and a fictitious domain method.

Hybrid Cell Model: * Lipid Bilayer: Modeled using a continuum approach based on the Helfrich bending energy, capturing the membrane's resistance to bending. * Cytoskeleton Network: Represented by a discrete model using worm-like chains (WLC) connected by springs, mimicking the viscoelastic properties of the cell's internal structural network. * Constraints: Local and global area constraints, along with volume constraints, are incorporated to ensure membrane incompressibility and maintain cell volume. * Mesh: A triangular mesh is used to represent the cell surface, with vertices and edges corresponding to physical components.

Fictitious Domain (FD) Method with Distributed Lagrange Multipliers (DLM): * Decoupling: A fractional step scheme is employed to decouple the fluid and solid equations into three sub-problems: fluid, solid, and Lagrange multiplier. * Fluid Solver: Uses a projection method for the incompressible Navier-Stokes equations. * Solid Solver: Employs a Newton-Krylov iterative method to solve the nonlinear equations of motion for the cell membrane. * Lagrange Multiplier: Used to enforce the no-slip boundary condition at the fluid-solid interface. * Level Set and ENO Reconstruction: These techniques are used to accurately compute the curvature and surface Laplacian of the cell membrane, which are crucial for calculating the bending forces. * Immersed Boundary Interpolation: Used to transfer information between the fluid and solid domains.

The novelty lies in the combination of the hybrid cell model (integrating continuum and discrete elements for membrane mechanics) with the DLM/FD formulation, enhanced by advanced numerical techniques like level set and ENO for accurate curvature calculations.

Datasets

Not discussed in the paper. The paper relies on existing experimental data for validation.

Key Insights and Ideas

  • Hybrid Model Efficacy: The hybrid model effectively captures the complex mechanical behavior of cell membranes by combining the strengths of continuum and discrete modeling approaches for the lipid bilayer and cytoskeleton, respectively.
  • Fictitious Domain Advantage: The fictitious domain method simplifies meshing for complex geometries, allowing for the use of structured grids even when simulating deformable, complex-shaped cells.
  • Curvature Calculation Accuracy: The integration of level set methods and ENO reconstruction is critical for accurately computing the high-order derivatives needed for membrane bending forces, a common challenge in such simulations.
  • Predictive Power: The method demonstrates the ability to accurately predict cell deformation and motion under various flow conditions, aligning well with experimental observations.

Key Results

  • Numerical Accuracy: The fluid solver demonstrates second-order accuracy, while the curvature and surface Laplacian computations achieve first-order accuracy.
  • RBC Stretching: The model accurately reproduces the axial and transverse diameters of red blood cells (RBCs) under stretching forces, showing good agreement with experimental data.
  • Tank-Treading Motion: The simulation successfully captures the tank-treading motion of RBCs in shear flow, with frequencies closely matching experimental results.
  • Platelet Flipping: The method accurately simulates the flipping dynamics of platelets in shear flow, aligning with Jeffery's orbit theory.
  • Bacterial Swimming: The model can simulate bacterial swimming, producing helical trajectories consistent with existing studies.

Limitations and Future Work

Limitations: * The paper acknowledges that the accuracy of curvature and surface Laplacian computation is first-order. * While the model is validated against experimental data, the computational cost for complex 3D simulations with very fine grids might still be significant.

Future Work: * The authors suggest that the developed model and numerical methods can be extended to study more complex fluid-cell interaction problems, such as cell adhesion, cell division, and interactions with complex microenvironments. * Further refinement of the numerical schemes for higher-order accuracy in curvature calculations could be explored. * Investigating the impact of different cell types and their unique membrane properties would be a natural extension.

Summary updated

BibTeX
@article{lin2015fictitious,
  author = {Guang Lin and Wenrui Hao and Zhiliang Xu and Chun Liu and Guang Lin},
  title = {{A fictitious domain method with a hybrid cell model for simulating motion of cells in fluid flow}},
  journal = {Journal of Computational Physics},
  year = {2015},
  doi = {10.1016/j.jcp.2014.09.020},
}
Figure