Model-based inference of cell cycle dynamics captures alterations of the DNA replication programme

The cell cycle of eukaryotic cells consists of several processes that must be carefully orchestrated and completed in a timely manner. Alterations in cell cycle dynamics have been linked to the onset of various diseases, underscoring the need for quantitative methods to analyze cell cycle progression. Here, using a combination of high-throughput experimental data and theoretical modeling, we develop RepliFlow, a model-based approach to infer cell cycle dynamics from flow cytometry data of DNA content in asynchronous cell populations. We model the DNA content distribution as the result of both noisy single-cell measurements and the populations age structure. We show that RepliFlow captures not only changes in the length of each cell cycle phase but also alterations in the underlying DNA replication dynamics. RepliFlow is species-agnostic and recapitulates results from more sophisticated analyses based on nucleotide incorporation. Finally, we propose a minimal DNA replication model that enables to derive microscopic insights about origin firing rates and replication fork speed from population-wide DNA content measurements. Our work presents a scalable framework for inferring cell cycle dynamics from flow cytometry data, enabling the characterization of replication program alterations.