Skip to main content
Figure 3 | Theoretical Biology and Medical Modelling

Figure 3

From: On the origins of the mitotic shift in proliferating cell layers

Figure 3

Computational predictions of the mitotic shift based on a stochastic model of the side-gaining process. (A) Prediction of the mitotic cell shape distribution for the Drosophila wing disc epithelium (empirical values shown in grey). In the absence of cleavage plane bias (blue), the computational prediction is not quite as accurate as when the bias is included (red). Sample sizes for the empirical Drosophila mitotic polygonal distribution are given under the heading “Sample sizes for overall and mitotic cell shape distributions” in the Methods section. (A’) Prediction of the mitotic cell shape distribution in the epidermis of Cucumis. Here, cleavage plane bias similarly improves the accuracy of the prediction. Sample sizes for the empirical Cucumis mitotic polygonal distribution are given under the heading “Sample sizes for overall and mitotic cell shape distributions” in the Methods section. (B) For the Drosophila prediction, a plot of the (l2-norm)2 deviation from the empirical mitotic cell shape distribution as a function of the relationship between division likelihood and polygon class. Here, division likelihood is assumed to increase exponentially as a function of polygon class. The ordinate (exponential constant) gives the precise form of the exponential. Note that positive values strongly outperform negative values. Hence, a model in which division likelihood increases with polygon class is more consistent with the data than a model in which it decreases or remains the same. (B’) For Cucumis, the results are nearly identical to those of Drosophila.

Back to article page