1. Data-driven models for cell motility in complex 2- and 3-dimensional environments
- Author
-
Scott, Marianne
- Abstract
Studying cell motility is of vital importance for health, for knowing how cells behave and are affected by, and can themselves cause, disease. Mathematical modelling of such behaviour has proved beneficial for furthering knowledge of important motility processes in many different cell types. This work aims to define and analyse data-integrated mathematical models for cell motility in 2 and 3 dimensions, specifically applied to glioblastoma tumour cells and surface-attached P. aeruginosa bacterial cells. Models are outlined, tested on in silico data, parametrized where possible and assumptions are studied in detail. As a result, recommendations are made for how subsequent data could be collected to further improve the prediction and validation of these models. A comprehensive framework is developed for the analysis of cell tracking data in 2 and 3 dimensions which allows a user to study various aspects of the Persistent Random Walk model as applied to these tracks, looking at speeds, persistence time, mean squared displacement and root mean squared speed. In silico simulations show good agreement with model predictions, however the model is incapable of describing the experimental data, as evidenced by lack of agreement in speed distributions and the speed parameter changing with time. A Bayesian approach to estimating these parameters is also considered, with estimates of persistence time seen to be inflated here compared to those from the frequentist approach. A newly-observed twiddling mechanism used in chemotaxis by P. aeruginosa is also studied, through rigorous hypothesis testing of assumptions about this motion. An individual-based model is employed to simulate the resulting chemotactic motion, which shows good agreement with results from the specified analytic model, though the model cannot currently be validated against experimental data due to lack of appropriate data for parameter estimation.
- Published
- 2021
- Full Text
- View/download PDF