Modelling Realistic Neuron Shape Development in А Realistic Cellular Tissue Environment
Jordan Genoff (Technical University of Sofia at Plovdiv)
Realistic neuron spatial shape development has proven to be a highly complex research topic in neurobiology. There exist several computational modeling methodologies with their mathematical foundations and software implementations for this subject. Though each of them approaches the problem in its individual and original way, they all have a basic common feature - they are neuron-centric, i.e. it is assumed that the neuron shape develops as a result of morphological changes mostly caused by intracellular processes. A significant number of these processes are initiated as reactions to extra-neuronal biochemical or mechanical impacts, but the latter two are usually considered in a generalized or, in the best case, probabilistic manner. The incontrovertible fact that neurons are not alone in space and their neighboring cells are not transparent at all is somehow disregarded.
This paper proposes the idea that the exact geometric configuration of the surrounding cells plays a highly important role in the specific shape development of every single neuron. A modeling approach is suggested that aims at detailed investigation of how neuron growth cones are supervised by the neighboring cells and their contact properties. A Cellular Potts Model (CPM) is chosen as a simulation implementation technique. However, which is another contribution in the paper, CPM is utilized in an unusually hard parameter and input data setup, which reveals some avoidable disadvantages of this model.
Preferred presentation format: