A fundamental aspect of the top-down model is a series of components for modeling the relationships between nuclei across time, allowing us to simultaneously predict and link the cell nuclei during the decoding phase.
This implies a homogenization mechanism, which offers a better number of nuclei with time [16].
The location of nucleation events was recorded as the number density of nuclei vs time.
An early attempt at lineage tracing, StarryNite, used a separate detection algorithm based on a modified watershed algorithm and a custom distance based algorithm to link the nuclei across time (Bao et al. 2006).
The absence of correlation means that for some nuclei and time points the impact from the two sites can be of the same sign, while for others of the opposite signs.
Another improved method, nuc-seq or single nucleus exome sequencing (SNES), has been developed to reduce the bias, this method combines flow-sorting of single G1/0 or G2/M nuclei, time-limited isothermal MDA, exome capture, and NGS [49, 50].
Even in the absence of RanGTP, Impβ binds more avidly to FG repeats and diminishes influx of transportin into nuclei ≈50-times more strongly than full-length importin β (Supplementary Figure S1).
Where the dynamic submodel considers the relationship between nuclei in adjacent time points, the time-to-event submodel considers the aggregate relationship of all cells within a branch across multiple time points.
The linkage component describes the fundamental element responsible for establishing relationships between nuclei across two time points.
Finally, TEM revealed strong nuclear condensation at 3 days compared to control nuclei at this time point.
As we describe, our method extends upon these approaches by developing models that describe specific constraints on how the nuclei change over time (e.g., size, location), and how they respond during division events (e.g., angle of division, division distance).
