The Project “Innovative mathematical modelling for cell mechanics: global approach from micro-scale models to experimental validation integrated by reinforcement learning” (Cell:GAMER) aims to develop an innovative approach using mathematical models and experimental validation, supported by reinforcement learning techniques, to understand nonlinear phenomena in cell mechanics such as adhesion and growth. These phenomena are crucial for fundamental biological processes like morphogenesis and pathogenesis. The project will employ a multidisciplinary approach to characterize mechanical, thermal, and chemical properties of the systems under scrutiny, requiring advanced mathematical techniques and instrumentation. It will be structured into different research lines, converging to integrate the approaches synergistically. The project will focus on discrete modeling at the microscale, finite element-based numerical methods, experimental validation of mechanical behavior, and the application of physics-based machine learning techniques to optimize model training and predictive capabilities.
The Project is structured in different research lines to achieve the following goals:
1- Development of discrete models at the microscale based on elements that exhibit multistability to study how the microscopic details are responsible for emergent phenomena at the meso and macroscale.
2- Modeling of the complex response of the cells’ systems across the scales using Finite Element-based numerical methods comparing the results with those derived from the models at the microscale.
3- Experimental validation on adhered cell cultures characterizing elastic properties, surface topography and adhesion phenomena of cells under specific load conditions to confirm the theoretical outcomes about the mechanical behavior.
4- Study of physics-based machine learning techniques to include information obtained from the developed models and the experimental validation and optimize the training phase to improve the output and predictive capabilities of these methods.
