(A) Histograms of cell-cycle lengths for a single ancestor cell and its progeny. S8 Fig: Results of long-term behavior predicted by the model. Significant difference between 10% FBS and 15% FBS is observed only when incomplete cell cycles are included. Kaplan-Meier curve maps division events onto time, including cells which divided only once, so that either their birth or death are not known. (E) Survival function mapping division events onto time, based on cells that divided at least twice during the experiment, so the time of birth and death could be estimated. (D) G1 phase takes proportionally less time under 10% FBS (Wilcoxon rank sum test), as it is shown in the boxplot (105 samples for 10% FBS, and the same number of randomly selected samples for 15% FBS). The division moments were selected using procedure described in Methods section.
Each line denotes one cell, black solid line is mean trace, black dashed line is median trace. Standard deviations were calculated using Monte Carlo cross validation and 10,000 iterations. (B) Pearson rank correlations between lengths of the phases and the cell cycle. Changes in the lengths of the cell cycle and of the S/G2/M phases are statistically significant (Wilcoxon rank sum test). Lower dose of serum causes extension of G1 and S/G2/M phases and as a consequence of the whole cell-cycle length. All distributions have characteristic lognormal-like shape additional information about medians are included directly on the plots. Histograms were normalized, the height of each bar is equal to the probability of selecting an observation within the corresponding bin interval, and the height of all of the bars sums up to 1. Calculations were performed based on 105 and 642 measurements for 10% and 15% of FBS, respectively. (A) Comparison of cell cycle, G1 phase and combined S/G2/M phases durations for two serum (FBS) concentrations. S7 Fig: Populations with different growth factors concentrations.
0 kommentar(er)
