Dagmar Iber

Lucas Daniel Wittwer, Roberto Croce, Sebastian Aland et al.

Organogenesis has been studied for decades, but fundamental questions regarding the control of growth and shape remain unsolved. We have recently shown that of all proposed mathematical models only ligand-receptor based Turing models successfully reproduce the experimentally determined growth fields of the embryonic lung and thus provide a mechanism for growth control during embryonic lung development. Turing models are based on at least two coupled non-linear reaction-diffusion equations. In case of the lung model, at least two distinct layers (mesenchyme and epithelium) need to be considered that express different components (ligand and receptor, respectively). The Arbitrary Lagrangian-Eulerian (ALE) method has previously been used to solve this Turing system on growing and deforming (branching) domains, where outgrowth occurs proportional to the strength of ligand-receptor signalling. However, the ALE method requires mesh deformations that eventually limit its use. Therefore, we incorporate the phase field method to simulate 3D embryonic lung branching with COMSOL.

Marius Almanstötter, Roman Vetter, Dagmar Iber

Parameter estimation for differential equations from measured data is an inverse problem prevalent across quantitative sciences. Physics-Informed Neural Networks (PINNs) have emerged as effective tools for solving such problems, especially with sparse measurements and incomplete system information. However, PINNs face convergence issues, stability problems, overfitting, and complex loss function design. Here we introduce PINNverse, a training paradigm that addresses these limitations by reformulating the learning process as a constrained differential optimization problem. This approach achieves a dynamic balance between data loss and differential equation residual loss during training while preventing overfitting. PINNverse combines the advantages of PINNs with the Modified Differential Method of Multipliers to enable convergence on any point on the Pareto front. We demonstrate robust and accurate parameter estimation from noisy data in four classical ODE and PDE models from physics and biology. Our method enables accurate parameter inference also when the forward problem is expensive to solve.