MCMC Techniques for Parameter Estimation of ODE Based Models in Systems Biology

摘要: Ordinary differential equation systems (ODEs) are frequently used for dynamical system modelling in many science fields such as economics, physics, engineering, and biology. A special challenge biology is that ODE typically contain kinetic rate parameters, which unknown have to be estimated from data. However, non-linearity of together with noise the data raise severe identifiability issues. Hence, Markov Chain Monte Carlo (MCMC) approaches been estimate posterior distributions parameters. designing a good MCMC sampler high dimensional multi-modal parameter remains challenging task. Here we performed systematic comparison different techniques this purpose using five public domain models. The included Metropolis-Hastings, parallel tempering MCMC, adaptive MCMC. In conclusion, found specifically produce superior estimates while benefitting inclusion our suggested informative Bayesian priors parameters variance.