Cell cycle models and mother-daughter correlation.

Among the different experimental features that are usually used to measure the variability of cell cycle duration (Tc), the mother-daughter correlation is a special case since apparently contradictory results have been reported from different experimental systems. Indeed, positive, null and negative experimental values have been found, raising problems in cell cycle modeling in as much as one would like a general, unified view of the control of the cell cycle. Such views are provided by mathematical models. We present a method to derive the value(s) of the correlation between mother and daughter Tcs predicted from any such model, provided Tc can be analytically expressed. An exact calculation is proposed for the case when Tc can be expressed as a linear combination of the parameters involved. Otherwise, using Taylor series expansion, an approximate expression for the correlation can be derived. In both cases, the predicted values can be compared with experimental ones. The proposed method is illustrated using the Growth Controlled model of Koch (1980, Nature 286: 80) and the Continuum model of Cooper (1982, J. theor. Biol. 94: 783). Whereas Koch's model leads to inflexible negative values of the correlation between mother and daughter Tcs, our results demonstrate that differences in the mean duration of the G1 phase may account for differences in the sign of this correlation in Cooper's model. Furthermore, the proposed method can also be applied to other correlations such as the correlation between sibling cell Tcs.(ABSTRACT TRUNCATED AT 250 WORDS)