Quantifying the organisation in cancer cell partitioning noise

Phenotypic heterogeneity in genetically identical cancer cell populations arises due to extrinsic and intrinsic sources of biological noise. In particular, partitioning noise, i.e. fluctuations in the cellular components during cellular division, is regarded as one of the major enhancers of phenotypic variability as such fluctuations can lead to dramatic changes in the cell state and thus to transitions between different phenotypes. The resulting plasticity increases the heterogeneity of tumors and changes their ability to resist treatments. Here, we show how to quantify the extent of fluctuations in the partitioning of different cellular components during the division of human leukemia cells. In particular, comparing the outcomes of a multi-fluorescent stain experiment with the predictions of a minimal statistical model, we found that a correlated binomial statistic explains the partition of cell cytoplasm and mitochondria together with their observed correlations. In perspective, our model can be applied to study the organisation in the partitioning of a wide range of cellular components paving the way to a better understanding of noise-driven cellular processes.