Servicenavigation

Project Details

Back

Projekt Print View

spBIGDATA_Mathematical and Physical modeling of Single Particle Tracking - Big Data approach

Subject Area Statistical Physics, Nonlinear Dynamics, Complex Systems, Soft and Fluid Matter, Biological Physics
Mathematics
Term from 2017 to 2023
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 380893586
 
Final Report Year 2023

Final Report Abstract

Modern microscopic techniques and supercomputing studies unveil the detailed motion of tracer particles in living biological cells, the motion of lipids and proteins in membranes, or the internal dynamics of bio-molecules. Often the observed diffusive motion deviates from the laws of normal Brownian motion (linear mean-squared displacement, Gaussian displacement probability density function). This emerging phenomenon is called anomalous diffusion. Since the 1960s a number of anomalous diffusion models have been formulated, including long-range correlated motion (typically observed in viscoelastic systems) or motion that is interrupted by scale-free trapping times (for instance, seen in internal protein dynamics or in the motion of membrane proteins). However, the increasing degree of detail as well as length of the measured trajectories requires ever more sophisticated stochastic models to accommodate the observed dynamics. Concurrently, from garnered data we want to obtain physical information about the observed system. To this end specific data-centric methods need to be developed. In the current project we combine the expertise of the Applied Maths group at the Wrocław University of Science and Technology and the Theoretical Physics group at the University of Potsdam, to design and study novel stochastic models for the analysis of single-particle trajectories in complex systems as well as develop new strategies to extract physical information and the relevant parameters from recorded time series. For both directions our Polish-German consortium delivered a broad range of different, and complementary approaches, including new anomalous diffusion models for disordered systems, new statistical observables, as well as dedicated computer-based data-centric approaches based on Bayesian statistics, feature-based machine learning, and deep-learning methods.

Publications

 
 

Additional Information

© 2025 DFG Disclaimer / Copyright / Privacy Policy

Textvergrößerung und Kontrastanpassung