Development and Application of Cell-Based Simulation Frameworks for Morphogenetic Problems
EMBARGOED UNTIL 2027-06-24
Loading...
Author / Producer
Date
2025
Publication Type
Doctoral Thesis
ETH Bibliography
yes
Citations
Altmetric
EMBARGOED UNTIL 2027-06-24
Data
Rights / License
Abstract
Morphogenesis is the developmental process by which organisms acquire their shapes. This complex process relies on the dynamic interplay between mechanical forces and the diffusion of signaling molecules, which work together to guide cellular behaviors, ultimately leading to the formation of functional organs. Studying these dynamics experimentally in living embryos is challenging, given their fragile and inaccessible nature. Even when all factors influencing their growth are known, their combined effects can remain elusive. In this context, computer simulations serve as a valuable supplement to experimental methods, enabling the artificial reproduction of biological tissues. By manipulating the development of these virtual tissues, deeper insights can be gained into the mechanisms driving morphogenesis. Even though the complex shapes of cells as well as entities such as the extracellular matrix play pivotal roles during the morphogenesis of tissues, these features are often simplified or omitted in simulations due to the computational complexity and cost of modeling them accurately. This thesis aims to address this gap by developing new computational methods to accurately simulate cellular behaviors and estimate cellular properties during tissue morphogenesis. The first part of the thesis is concerned with the development of a novel high-resolution 2D model, named PolyHoop, capable of simulating soft particles such as biological cells, bubbles, and emulsions. The versatility of the designed model is showcased by simulating a wide range of phenomena. The numerical stability and efficiency of PolyHoop are then demonstrated by simulating the growth of a tissue from one to a million cells. The second part of this thesis discusses how the approach that underpins the efficiency and versatility of PolyHoop has been extended into three dimensions in a novel model called SimuCell3D. Similar to PolyHoop, SimuCell3D can simulate biological tissues at cellular resolution with high geometric fidelity. Its efficient implementation makes it several orders of magnitude faster than previously published implementations. First are presented the various features natively incorporated in SimuCell3D such as automatic cell surface polarization, organelles, or local mesh refinement. SimuCell3D is then applied to study the relationship between cell mechanical parameters and the architecture of a tissue as well as the shape of cells in pseudostratified tissues. The third part of this thesis discusses how a simplified version of SimuCell3D has been combined with a parameter optimization technique to infer the mechanical properties of cells in imaged tissues. This method, called OptiCell3D, uses short, fully differentiable simulations to determine the optimal combination of cell pressures and cortical tensions needed to keep a tissue in equilibrium. An extensive benchmark of this novel approach is performed to demonstrate its superior accuracy compared to previously published methods. The final part of the thesis is concerned with the study of the interactions between the different cell types of the developing pancreas. The gene expression profiles of isolated single cells are used to identify potential interactions between the major cell types of the pancreas. The devised analysis predicted a total of more than 40,000 potential interactions. Several of these predictions were experimentally validated, revealing previously unknown chemical signaling interactions that influence pancreatic morphogenesis. The identified interactions can now be integrated into the mechanical models developed in this thesis, enabling an accurate representation of both cell mechanics and chemical signaling during pancreatic morphogenesis.
Permanent link
Publication status
published
External links
Editor
Contributors
Book title
Journal / series
Volume
Pages / Article No.
Publisher
ETH Zurich
Event
Edition / version
Methods
Software
Geographic location
Date collected
Date created
Subject
Biological Tissue Simulation
Organisational unit
03791 - Iber, Dagmar / Iber, Dagmar