BPCPPDYSOE21 – scientific programme

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

BP: Fachverband Biologische Physik

BP 11: Poster A: Single Molecule, Multicellular, Bioimaging, Focus Sessions, etc.

BP 11.25: Poster

Monday, March 22, 2021, 16:30–19:00, BPp

Determination of the effective adhesion parameter for the sorting behavior of a cell system with several cell types using statistical learning methods — •Philipp Roßbach, Steffen Lange, Hans-Joachim Böhme, and Anja Voss-Böhme — Hochschule für Technik und Wirtschaft Dresden

The process of cell sorting plays an essential role in development and maintenance of tissues. To understand this process, mathematical modeling can assist cell biological research by providing means to analyze the consequences of different hypotheses on the underlying mechanisms. In the Differential Adhesion Hypothesis (DAH) by Steinberg (1962) it is assumed that cell sorting is determined by quantitative differences in cell type specific intercellular adhesion strengths. An implementation of the DAH is the Differential Migration Model (DMM) by Voss-Böhme and Deutsch (2010). From this DMM an effective adhesion parameter (EAP) for systems with two cell types can be derived analytically which predicts the asymptotic sorting pattern. However, the existence and form of such an parameter for more than two cell types is unclear.

Here, we investigate numerically the existence of an EAP for systems with more than two cell types. We rely on in-silico time-series data that is produced by a cellular automaton which emulates the DMM and classify the segregation behavior using statistical learning methods such as SVM and Logit Model. We use these tools to demonstrate the existence of an EAP for three cell types which matches our analytical prediction for systems with arbitrary many cell types.