Parametric modeling of cellular state transitions as measured with flow cytometry

Gradual or sudden transitions among different states as exhibited by cell populations in a sample under particular conditions or stimuli can be detected and profiled by flow cytometric time-course data. Such temporal profiles however often contain non-Gaussian features due to transient states, and thus present unique modeling challenges. We propose a precise and parametric modeling method based on finite mixtures of skew t-Normal distributions that are robust against non-Gaussian features caused by asymmetry and outliers in data. Further, we present a new greedy EM algorithm for fast and optimal model selection. The parsimonious approach of our greedy algorithm allows us to detect the genuine dynamic variation in the key features as and when they appear in time course data. Thus our model parameters can provide precise characteristics of cellular state transition. We applied our method to learn the temporal features of yeast cell cycle progression based on knockout of S-phase triggering cyclins Clb5 and Clb6, and to statistically compare the delay phenotypes due to differential regulation of the two cyclins.