Optimization of E.Coli heat shock response parameter tuning using distributed and integrated genetic algorithms

The progress of computer made the analysis that was difficult in traditional technology possible. In Gene Regulatory Network Simulation to recreate and understand actual biological reaction, we build a reaction model and find the several unknown parameters in order to show the same behavior as actual. However it is difficult to optimize them because the scopes of them are logarithm scale wide and the fitness space is multidimensional and multimodal. We optimized it using Distributed Genetic Algorithms (DGA) which has multiple populations. DGA showed better performance, if and only if the parameter of the emigrating operation is appropriate. Therefore we propose Distributed and Integrated Genetic Algorithms (DIGA) to reduce the cost to find the appropriate parameters of operation. We carried out some optimizing experiments on parameter tuning of Heat Shock Response simulation of E. Coli and several mathematical functions and inspected the characteristic of DIGA.