Skip to main content
Fig. 4 | EvoDevo

Fig. 4

From: Around the clock: gradient shape and noise impact the evolution of oscillatory segmentation dynamics

Fig. 4

Comparison of genome, network and oscillatory dynamics properties. a Boxplot of the morphogen level at which individuals reach a stable expression (after the transition from the oscillatory to the non-oscillatory regime). b Violin plots (vertical histogram) of the number of genes and transcription factor binding sites (TFBS) in the pruned genomes of shallow-gradient (dark) and steep-gradient (light) simulations. Dots indicate the median value. Mann–Whitney U test between shallow and steep: genes, \(p=0.006\); TFBS, \(p=0.0007\). After removing the 14 largest genomes from both sets: genes, \(p=0.003\); TFBS, \(p=0.0001\) (corrected for ties with jitter). c Violin plots of the number of positive and negative feedback loops in the pruned networks of the shallow- and steep-gradient simulations. (MW test: pos.FBL, \(p=0.005\); neg.FBL, \(p=0.0003\). After removing 14 genomes with most loops: pos.FBL, \(p=0.008\); neg.FBL, \(p=0.0001\)). d Histogram of the number of loops (FFL and FBL) of a certain size. All histograms of individual simulations have been summed for this average histogram. e Histogram displaying for all successful individuals their frequency difference between oscillations in the growth zone and at the end of the profile, before sustained oscillations cease. (see indication in the profile on the left: a nice example of a strongly sloped frequency profile with a large difference). Profiles to the right of the red line are classified as “sloped” in Table 2. Note that the damped oscillators are grouped in the bin with 0.0 frequency difference. Bin size: 0.01

Back to article page