Computer models of spatially-patterning gene networks

Computer models of spatially-patterning gene networks

Enough is known about the genetics of limb development to be sure that it involves many signalling molecules (Shh, BMPs, FGFs, Wnts) and many transcription factors (Msx, Hox, Meis) which are wired together into a complex gene network. We aim to explore how these networks function within the computer model – both in control of the cell behaviours that govren the limb bud shape, and also in another famous patterning case: the spatial organisation of the skeletal elements.

For skeletal patterning, different patterning strategies have been proposed within the literature to explain the process ranging from pure “Turing-type” reaction-diffusion models at one extreme, to morphogene gradient models on the other. We are exploring the ways that these different principles might work at the detailed level of gene regulatory networks. We have written simulation software which allows the spatial-patterning capabilities of millions of different gene networks to be explored. Currently the model is 1-dimensional, but will soon be extended to 3D and placed within the context of the normal limb morphology. In parallel with this, we are using OPT and confocal microscopy to gather accurate 3D expression data on genes thought to be involved in the initiation of mesenchymal condensation. Our goal will then be to test existing hypotheses and explore new ideas about how this particular example of pattern formation is controlled.