Extended Data Fig. 5: Survival curves for different organoids.

a) Kaplan-Meier curves describing the fraction of non-divided cells as a function of time since division, for 5 organoids where manual reference data was available for the complete crypt (grey curve). The shaded region denotes the 95% confidence interval. b) Kaplan-Meier curves describing the fraction of non-divided cells as a function of the time since the division of the sister cell. c) Comparison of manually annotated data to automated analysis for the estimation of three key parameters of lineage dynamics. Color indicates whether the variation in cell cycle duration or the probability that the cell will divide again was calculated for all cell cycles (red) or for cell cycles relative to the moment of division of the sister cell (blue). Cell cycle times show almost perfect correlation between manual and automated data. The deviation in the cell cycle time is more sensitive to outliers and consequently shows poorer correlation. The non-dividing fraction again shows strong correlation. d) Overlay of all Kaplan-Meier curves of automatically tracked organoids, n = 20, which describes the fraction of non-divided cells as a function of time since division. Black line represents the combined data. e) Same but as a function of time since the sister division. f) Statistical analysis of organoid lineage parameters. Error bars indicate standard deviations around the mean. Cell cycle times only show limited variation across organoids and experiments, with standard deviation <10% of the mean cell cycle time (left panel). The variation around the cell cycle mean for cell cycle times is significantly larger than the variation between the cell cycle times of sisters (middle panel), suggesting strong correlation between sister pairs. This is further supported by the results that given that the sister divides, the non-dividing fraction becomes close to zero instead of around 30% (right panel).