Extended Data Fig. 2: Identification of the typical cell-cycle phase of karyokinesis, which is assumed to denote the end of DNA replication.

(a) Microscopy image segmentation to locate the nucleus in the mother cell containing mRFP1 fused to the histone protein Hta2, and to calculate the abundance of the fusion. Details of automatic nucleus segmentation are provided in Methods. To calculate the abundance of the fusion Hta2-mRFP in the mother cell nucleus, we sum the intensities of the pixels located within the segmented nucleus. This experiment was performed once, however, similar nucleus segmentation micrographs were observed in all microscopy time points in 105 cell cycles. (b) The relative abundance of the fusion Hta2-mRFP1 in the mother-cell nucleus during the cell cycle to identify the phase of karyokinesis. In each individual cell-cycle trace, we divide the abundance by the value at the phase zero (the first mitotic exit ME). The circular markers correspond to different cell-cycle traces and phases. The plot is built analogously to Fig. 1a. Cell cycles whose duration is smaller or equal to 120 mins are considered for this analysis. The average cell-cycle duration measured in minutes is indicated. To summarize the behavior of the relative abundance during the cell cycle, we obtain the Gaussian process (GP) regression, whose mean and region of high posterior probability density (mean ± SD) are presented as the thick curve and the shaded area. The radial basis function (RBF) kernel assuming smoothness with the length scale range [0.05, 0.1] and the white kernel explaining the noise in the data with free noise level are used as a prior. We identified the typical cell cycle phase of karyokinesis as the middle time point between the maximal and minimal values of the GP regression mean after budding (the black vertical line).