Modeling of Collective Cell Behaviors Interacting With Extracellular Matrix Using Dual Faceted Linearization
Name
v001t14a005-dscc2018-9164.pdf
Description
Published version
Size
1.38 MB
Format
Adobe PDF
Checksum (MD5)
1912cd3763b33c4de57b3ddffdbd53c5
Author(s)
Mayalu, Michaëlle N.
Kim, Min-Cheol
Asada, Haruhiko
Date Issued
September 2018
Journal
ASME 2018 Dynamic Systems and Control Conference
Publisher
American Society of Mechanical Engineers
Citation
Mayalu, Michaëlle N. et al. "Modeling of Collective Cell Behaviors Interacting With Extracellular Matrix Using Dual Faceted Linearization." ASME 2018 Dynamic Systems and Control Conference, September 2018, Atanta, Georgia, USA, American Society of Mechanical Engineers, September 2018 © 2018 ASME
Version
Final published version
Abstract
Cells interacting over an extracellular matrix (ECM) exhibit emergent behaviors, which are often observably different from single-cell dynamics. Fibroblasts embedded in a 3-D ECM, for example, compact the surrounding gel and generate an anisotropic strain field, which cannot be observed in single cell-induced gel compaction. This emergent matrix behavior results from collective intracellular mechanical interaction and is crucial to explain the large deformations and mechanical tensions that occur during embryogenesis, tissue development and wound healing. Prediction of multi-cellular interactions entails nonlinear dynamic simulation, which is prohibitively complex to compute using first principles especially as the number of cells increase. Here, we introduce a new methodology for predicting nonlinear behaviors of multiple cells interacting mechanically through a 3D ECM. In the proposed method, we first apply Dual-Faceted Linearization to nonlinear dynamic systems describing cell/matrix behavior. Using this unique linearization method, the original nonlinear state equations can be expressed with a pair of linear dynamic equations by augmenting the independent state variables with auxiliary variables which are nonlinearly dependent on the original states. Furthermore, we can find a reduced order latent space representation of the dynamic equations by orthogonal projection onto the basis of a lower dimensional linear manifold within the augmented variable space. Once converted to latent variable equations, we superpose multiple dynamic systems to predict their collective behaviors. The method is computationally efficient and accurate as demonstrated through its application for prediction of emergent cell induced ECM compaction.
MIT Department
Massachusetts Institute of Technology. Department of Mechanical Engineering
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1115/dscc2018-9164
