Regensburg 2022 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
DY: Fachverband Dynamik und Statistische Physik
DY 40: Brownian Motion and Anomalous Diffusion
DY 40.4: Vortrag
Donnerstag, 8. September 2022, 10:45–11:00, H20
Size matters for Bayesian chemotaxis — •Julian Rode1, Maja Novak1,2, and Benjamin M. Friedrich1,3 — 1cfaed, TU Dresden, Germany — 2Department of Physics, University of Zagreb, Croatia — 3PoL, TU Dresden, Germany
Navigation by chemical cues, e.g., chemotaxis, is employed by single biological cells and animals. The size and speed of search agents dictate noise levels and thus optimal strategies to find a target.
Here, we address information theory of gradient sensing for an ideal agent and ask for optimal strategies as a baseline for real agents. We extend the seminal work on infotaxis [1], by applying its idea of maxi- mizing information gain to agents of finite size. These agents can now measure gradients both by temporal comparison due to their active motion [1], and by spatial comparison across their diameter, prompt- ing an optimal weighting of both information sources [2].
In the absence of noise, trajectories show stereotypic behavior; the entropy of directional uncertainty collapses onto a master curve pa- rameterized by a signal-to-noise ratio. Unlike [1], we account for rota- tional diffusion, which is prevalent for microscopic agents: Its compe- tition with information gain due to spatial comparison sets an effec- tive measurement time (given by the inverse geometric mean of a rate constant of information gain and the rotational diffusion coefficient), which is different from the typical bound argued by Howard Berg for bacterial chemotaxis (inverse rotational diffusion coefficient) [3].
[1] M. Vergassola et al, Nature (2007); [2] A. Auconi et al., EPL, in press (arXiv:2111.09630); [3] M. Novak et al., New J Phys (2021).