Planning blocked mitosis experiments for efficient estimation of population-doubling time and cell-cycle time

In this paper the standard errors of both classical and recently proposed estimators of mean population-doubling time and mean cell-cycle time are derived as functions of the number of cells counted and the counting times in a blocked mitosis experiment. Practical suggestions are made regarding the allocation of cell counts that will reduce the standard errors. It is shown that the proposal of Jagers (1977, Bulletin of the International Statistical Institute 47 (2), 418-422) for estimating cell-cycle time yields, quite generally, smaller standard error than do the classical estimators. Finally, a relationship between the mean and variance of cell-cycle time and the mean doubling time is derived; in some cases this relationship justifies the classical identification of mean doubling time and mean cycle time.