Dresden 2014 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

BP: Fachverband Biologische Physik

BP 8: Posters: Active cell and tissue mechanics

BP 8.6: Poster

Monday, March 31, 2014, 17:30–19:30, P3

Uncovering the slow mode dynamics of migrating tumor cells — •Christoph Mark, Claus Metzner, Julian Steinwachs, and Ben Fabry — Biophysics Group, University of Erlangen

Cell migration is usually analyzed by statistical methods that assume temporal homogeneity of the random process. Using long-time migration data from MDA-MB-231 tumor cells on 2D surfaces and in 3D collagen matrices, we show that this random walk process is highly inhomogeneous in time and accross the ensemble. Analysing the cell trajectories with a Hidden Markov Model reveals distinct migration modes with small mode switching rates. We thus describe the measured velocity time series v_t^→ as a first-order autoregressive process v_t+1^→=q_tv_t^→+σ_tg_t^→, in which the correlation factor q_t and the noise amplitude σ_t are treated as slowly varying superstatistical parameters. We show that the joint distribution p(q,σ) and the temporal correlations of these super-parameters provide a characteristic fingerprint of the cell’s migration mechanisms in different environments, whereas traditional time-averaging measures, such as the mean squared displacement, mask such differences almost entirely. An inhomogeneous model, based on the extracted superstatistics can reproduce a large set of characteristic data features quantitively.