MINSUK, S.B.; Konrad Lorenz Institute, Altenberg, Austria: First steps toward a generalized model cell for evo-devo computer simulations

Developmental simulations usually model a few cell types in a simple tissue, and conform to details of specific organisms. Simulated evolution must model cells and tissues, but proceed at the level of organisms and populations in environments. It should be general and open-ended. My goal is an open-ended model of a population of simple multicellular organisms with ontogenies. In this first step I lay the groundwork: a mechanically realistic (but fast executing) model of a generalized cell. I begin by adopting the modified Potts model, a fast, simple, and general model of passive cell movement and shape change via small random cell surface fluctuations, biased toward energetically favorable configurations (defined in terms of cell adhesion), within the limits of volume and surface area constraints. These constraints act as a computationally efficient stand-in for the explicit calculation of membrane tensions, cytoskeletal, and hydrostatic forces. I am adding two new features to the original model. First, an explicit cell membrane representation is required as the locus of cell-environment interaction. Second, although Potts model cells have arbitrary shape, they are homogeneous – precluding cell polarity, specialized local regions, or dynamic cell surface events. Adding heterogeneous cell surfaces will make these features possible, and increase the generality of the model. Finally, although the model works well for random, passive cell motion, previous approaches to adding active, directed motion have introduced an anomaly: they imply traction on an implicit rigid substrate, and behave incorrectly for motion in fluids or in the interior of cell aggregates. Adding heterogeneous membranes will make possible a more general implementation of cell crawling, in which cells exert traction directly on neighboring cells.