351

---

Click the image below to view the full size version of this cover.

Issue 351: Mathematical models of cancer evolution widely assume that population growth rate in the presence of drugs is a consequence of drug concentration-dependent single-cell division and death kinetics. In this study, the authors show that this assumption is not necessarily true in two cancer cell lines. By measuring single-cell division and death statistics in time-lapse microscopy experiments and applying Bayesian inference techniques, they demonstrate the inaccuracy of population dynamics predicted from the single-cell kinetics. They show that at the heart of this contradiction is the observation, made possible by lineage tracking, that closely related cells show similar fates post drug. Developing a simple stochastic model, they show that at approximately IC50 drug concentration a large fraction of the cell fate decisions happen well before drug administration. The drug modulates these decisions only to a small extent, which is quantifiable via our model. The study also demonstrates how analysing barcode diversity pre and post drug can be misleading and highlight the importance of measuring lineage correlations for timing fate decisions, and consequently developing relevant models of cancer cell proliferation. The figure was generated using a combination of ChatGPT-5 and Inkscape.