Parameter estimation using Improved Differential Evolution (IDE) and bacterial foraging algorithm to model tyrosine production in Mus Musculus (Mouse)

Abstract

The hybrid of Differential Evolution algorithm with Kalman Filtering and Bacterial Foraging algorithm is a novel global optimization method that is implemented in this research to obtain the best kinetic parameter value. The proposed algorithm is then used to model tyrosine production in mus musculus (mouse) by using a dataset, JAK/STAT (Janus Kinase Signal Transducer and Activator of Transcription) signal transduction pathway. Global optimization is a method to identify the optimal kinetic parameter using ordinary differential equation. From the ordinary parameter of biomathematical field, there are many unknown parameters and commonly the parameters are in nonlinear form. Global optimization method includes differential evolution algorithm which will be used in this research. Kalman Filter and Bacterial Foraging algorithm help in handling noise data and faster convergences respectively in the conventional Differential Evolution. The results from this experiment show estimatedly optimal kinetic parameters values, shorter computation time, and better accuracy of simulated results compared with other estimation algorithms. © Springer-Verlag 2013.