Computational modelling of retinal ganglion cell development

Individual retinal cell types exhibit semi-regular spatial patterns called retinal mosaics. These mosaics, enabling uniform sampling of visual information, are formed to varying degrees across cell subtypes. Retinal ganglion cells (RGC), amacrine, horizontal and photoreceptor cells are known to exhibit such layouts. In addition to cell body mosaics, dendritic arbours also form mosaics - dendrites of homotypic cells exhibit avoidance while dendrites of different types overlap.

Mechanisms responsible for the formation of such organised structures are still not well understood and they follow three main theories. (1) Homotypic cells prevent nearby cells from adopting the same type. (2) Cell tangential migration (CM), with homotypic cells repulsing each other. (3) Cell death (CD), with specific cell types (mainly RGCs) exhibiting high rates of apoptosis, increasing spatial regularity.

Here, we use BioDynaMo, an agent-based simulation framework, to build a detailed and mechanistic computational model of mosaic formation in 3D physical space. In particular, we investigate the implications of the three theories and their combinations, then compare them to experimental data we obtained in mouse.

We show that the CM mechanism yields the most regular mosaics. Moreover, the longest cellular migration distance achieved in these simulations is in agreement with experimental observations. We also found that CD can create regular mosaics only if the death rate is kept between 25 to 55%. After half of cells have died, CD appears to have a negative impact on regularity index, which matches our experimental findings.