|
Figure 8—figure supplement 1. Related to Figure 8. (A) Predictability of CDK activity on cell fate over time since anaphase. To evaluate the predictability of CDK activity (readout as the ratio of cytoplasmic-to-nuclear intensity of the CDK sensor) on proliferative vs. quiescent cell fate in different cell-cycle phases in C. elegans, we created a Receiver Operating Characteristic (ROC) curve for CDK activity at each time point relative to anaphase. Using the perfcurve function in MATLAB, we calculated the area under the curve (AUC) as the indicator of predictability. (B) Example ROCs that are used to calculate AUC in (A). (C) A histogram of all data points aggregated ≥50 min after anaphase shows that DHB ratios follow a bimodal distribution, suggesting the presence of two populations. (D) Raw data for each cell type/lineage are grouped together and mean time courses generated. (E) The area between these mean trendlines is then calculated. (D’) The assignment between experimental condition groups is randomized, mean time courses are calculated, and (E’) the area between the curves is calculated and recorded (purple line in F’). The randomization procedure, (D’) and (E’), is repeated 100,000,000 times generating a simulated distribution of the difference statistic: (F’). The true value of the difference statistic is compared to the distribution in (F’) generating a p-value (F) corresponding to the proportion of values more extreme than the observed difference statistic (i.e. the number of simulations in the magenta region, divided by the total number of simulations (108)).