Modeling Immune Cell Trajectories to Uncover Underlying Motility Drivers

Immune cells observed under a microscope often exhibit motion that deviates from simple random (Brownian) walks, yet the precise factors driving these deviations remain poorly understood. Statistical approaches, such as hidden Markov models, segment cell trajectories into multiple pure Brownian motion regimes and estimate a diffusion coefficient for each. While these methods provide insight into short-term movement changes and can be used to predict future positions, they offer limited explanation of the underlying biological motivation. Here, we propose a novel framework based on a transport equation to probe the fundamental causes of anomalous diffusion in immune cell trajectories. By fitting a generalized model to data, we can distinguish whether deviations from pure diffusion arise from a correlation in the cell’s motion—implying a “memory” of previous steps—or from a bias driven by an external factor, such as a cytokine gradient. In the latter scenario, immune cells may adjust their paths when approaching target cells (e.g., cancer cells), giving rise to a biased random walk. Moreover, the goal is to come up with a framework that accomodates the possibility of both memory effects and bias (a biased correlated random walk), enabling a more comprehensive description of immune cell motility. From our initial tests, it looks like this transport-based approach can spot unique signs of correlation or bias in immune cell movement just by analyzing time-lapse data. This could help us better understand how immune cells navigate their environments and may even open new avenues for guiding their behavior in therapeutic settings.

Modeling Immune Cell Trajectories to Uncover Underlying Motility Drivers

Monday, July 14 at 6:00pm in

SMB2025 Follow

Fatemeh Saghafifar

"Modeling Immune Cell Trajectories to Uncover Underlying Motility Drivers"

#SMB2025 Follow