Epithelial-mesenchymal transition (EMT) is a fundamental biological process that plays a central role in embryonic development, tissue regeneration, and cancer metastasis. Transforming growth factor-beta (TGF-beta) is a major and potent inducer of this cellular transition, which is comprised of transitions from an epithelial state to an intermediate or partial EMT state, then to a mesenchymal state. Using computational models to predict state transitions in a specific experiment is inherently difficult for many reasons, including model parameter uncertainty and the error associated with experimental observations. In this study, we demonstrate that a data-assimilation approach using an ensemble Kalman filter, which combines limited noisy observations with predictions from a computational model of TGF-beta-induced EMT, can reconstruct the cell state and predict the timing of state transitions. We used our approach in proof-of-concept "synthetic" in silico experiments, in which experimental observations were produced from a known computational model with the addition of noise. We mimic parameter uncertainty in in vitro experiments by incorporating model error that shifts the TGF-beta doses associated with the state transitions. We performed synthetic experiments for a wide range of TGF-beta doses to investigate different cell steady state conditions, and we conducted a parameter study varying several properties of the data-assimilation approach, including the time interval between observations, and incorporating multiplicative inflation, a technique to compensate for underestimation of the model uncertainty and mitigate the influence of model error. We find that cell state can be successfully reconstructed in synthetic experiments, even in the setting of model error, when experimental observations are performed at a sufficiently short time interval and incorporate multiplicative inflation. Our study demonstrates a feasible proof-of-concept for a data assimilation approach to forecasting the fate of cells undergoing EMT.

Cell fate forecasting: a data assimilation approach to predict epithelial-mesenchymal transition