Deciphering how natural selection emerges from demographic differences among genotypes, and reciprocally how evolution affects population dynamics, is key to understanding population responses to environmental stress. This is especially true in non-trivial ecological scenarios, such as programmed cell death (PCD) in unicellular organisms, which can lead to massive population decline in response to stress. To understand how selection may operate on this trait, we exposed monocultures and mixtures of two closely related strains of the microalga Dunalliela salina, one of which induces PCD, to multiple cycles of hyper-osmotic shocks, and tracked demography and selection throughout. Population dynamics were consistent between mixtures and monocultures, suggesting that selection on PCD does not involve strong ecological interactions. The PCD-inducing strain was maintained throughout the experiment despite an initial decline, by a combination of fast population rebound following each decline, and density-dependent competition dynamics near the stationary phase that were independent of these initial population fluctuations. As result of PCD maintenance, population decline in response to environmental stress was not counter-selected in our experiment, but persisted over 13 cycles of salinity. Our results highlight how analysing the demographic underpinnings of fitness and competition can shed light on the mechanisms underlying selection and eco-evolutionary dynamics.

This zenodo repository corresponds to the article "Linking selection to demography in experimental evolution of active death in a unicellular organism", published in Proceedings of the Royal Society B, Article DOI: 10.1098/rspb.2024.1707 . This study reports on our experimental explorations of eco-evolutionary dynamics in an puzzling context: active cell death leading to population decline in response to environmental stress.We tracked the population dynamics of two closely related strains of the halotolerant microalga Dunaliella salina (they differ in whether or not they trigger programmed cell death), in competition over 13 cycles of salinity rise (about 180 generations). This repositeroy contains:'code' folder : the entire workflow with the Rmarkdown scripts of analysis, intermediate inputs and outputs.'data_cytometer' folder : the raw data of the cytometer counts.'simulated_data_sets' folder : the 1000 simulated data sets.