Distributed parameter identification for a label-structured cell population dynamics model using CFSE histogram...

Area of Science:

• Mathematical Biology
• Computational Biology
• Cell Dynamics Modeling

Background:

• Cell population dynamics are crucial for understanding biological systems.
• Accurate parameter identification is essential for reliable model predictions.
• Experimental data on cell kinetics, using fluorescence markers, provides valuable insights.

Purpose of the Study:

• To develop a robust computational method for identifying unknown parameters in a cell population dynamics model.
• To focus on estimating the cell birth rate as a function of marker intensity.
• To address the challenges of ill-posed inverse problems in parameter estimation.

Main Methods:

• Formulation of a first-order hyperbolic PDE model for cell distribution based on fluorescence marker intensity.
• Parameterization of the cell birth rate using cubic Hermite splines.
• Application of a maximum likelihood approach for parameter estimation.
• Implementation of Tikhonov regularization to handle ill-posedness, with parameter selection via the discrepancy principle.

Main Results:

• A computational approach for robust identification of cell population model parameters was developed.
• The cell birth rate function was successfully parameterized and estimated.
• Tikhonov regularization effectively addressed the ill-posed nature of the inverse problem.
• The regularized parameter estimation yielded results consistent with experimental data, within measurement noise levels.

Conclusions:

• The developed computational method provides a robust framework for parameter identification in cell population dynamics.
• Accurate estimation of cell birth rates is achievable even with complex, ill-posed inverse problems.
• The findings demonstrate the utility of fluorescence-based kinetic data and advanced computational techniques in quantitative cell biology.