Melanoma Cell Colony Expansion Parameters Revealed by Approximate Bayesian Computation

Area of Science:

• Computational Biology
• Mathematical Modeling
• Cancer Research

Background:

• In vitro studies and mathematical models are crucial for understanding cell colony expansion, vital for cancer research.
• Estimating model parameters from experimental image-based data remains a significant challenge in this field.

Purpose of the Study:

• To develop and validate a novel approximate Bayesian computation (ABC) algorithm for parameter estimation in melanoma cell colony expansion models.
• To precisely infer key parameters like cell diffusivity (D), proliferation rate (λ), and cell-to-cell adhesion (q) using experimental in vitro data.

Main Methods:

• Development of a new approximate Bayesian computation (ABC) algorithm tailored for in vitro image-based data.
• Application of the ABC algorithm to estimate parameters governing melanoma cell (MM127) colony expansion under different experimental conditions (with/without chemical treatment).

Main Results:

• The ABC algorithm precisely inferred all parameters (D, λ, q) with small coefficients of variation (2-12%), even with limited prior biological knowledge.
• Posterior distributions of diffusivity (D) and adhesion (q) were found to depend on experimental elapsed time and initial cell density, while proliferation rate (λ) did not.
• Combining data from two experiments using the ABC approach enhanced the precision of D and λ estimates.

Conclusions:

• The proposed ABC algorithm offers a robust and precise method for estimating parameters in cell colony expansion models from experimental data.
• This approach advances our ability to understand and potentially predict cancer formation and progression by refining mathematical models with empirical evidence.