Dresden 2026 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
BP: Fachverband Biologische Physik
BP 14: Poster Session II
BP 14.48: Poster
Dienstag, 10. März 2026, 18:00–21:00, P2
Evolution-inspired exploration of pattern formation in reaction diffusion systems — •Ming Hong Lui1,2 and Erwin Frey1,2 — 1Faculty of Physics, Ludwig Maximilian University of Munich, Germany — 2Max Planck Schools Matter to Life, Max Planck Institute for Medical Research, Heidelberg, Germany
Protein reaction networks often contain complex network topology. Modelling them involves numerous variables such as concentrations, diffusion rates and kinetic coefficients. Each variable adds an additional dimension to the phase space that made sweeping parameters challenging and impedes comprehensive understanding of the full dynamics by demanding assumptions or simplifications of biological systems.
Here, we present an alternative pipeline to full parameter sweeps, to characterize pattern-forming regimes, by applying an evolutionary algorithm. We iteratively bootstrapped parameter combinations that demonstrated instability using linear stability analysis. Together with applying mutations, represented by random displacements on the high-dimensional phase space, we could focus our attention in the vicinity of instabilities to reconstruct a heuristic phase diagram more effective than evenly-spaced parameter sweeps.
We shall demonstrate how this algorithm can be applied to various reaction-diffusion models, where it provided more informed choices of diverse systems to run full numerical simulations on, bridging the gap bewteen analytical and numerical solutions.
Keywords: Pattern formation; Reaction-diffusion system; Evolution