Inferring the Stochastic Dynamics of Bacterial Growth and Shape Fluctuations

Growing rod-shaped bacterial cells exhibit stochastic shape dynamics but the nature of noise and fluctuations remains poorly understood. A quantitative understanding of the noise in bacterial growth and how it changes under different nutrient conditions would provide better insight into cell-to-cell variability and intergenerational fluctuations. Using multigenerational growth and shape data on single Escherichia coli cells, we derive the equations for cell growth and morphodynamics. Interestingly, we find that E. coli cells elongate faster than an exponential within an individual generation. We propose a physical model that explains this phenomenon based on heterogeneous patterning of cell growth. Using this new model and statistical inference on large datasets, we construct the Langevin equations for the shape dynamics of E. coli cells in different growth conditions. We then use the model to make predictions about bacterial adaptation to nutrient shifts, resulting in a simulation that accurately represents noise and fluctuations in cell shape in non-steady growth conditions.