| With the development of the big data,from the perspective of system,computational biology more provide research the possibility of quantitative analysis and qualitative analysis of complex systems.From the perspective of the data,in order to know more about the characteristics of complex systems,a lot of literature put forward with some appropriate mathematical model to describe and analyze the nature of system.Then,the researchers found that the model established,not only reduce the gap between experimental level and system level,but see some complex system process as important dynamic principle.And the accurate identification of parameters in model is the key to describe many complex process of system.As a result,more and more researchers pay more attention to the design of parameter identification algorithm.Parameter identification for biochemical reaction systems is one of the most challenging problems in the research of computational systems biology.In general,most algorithms for parameter identification encounter two important issues,i.e.high time complexity and low accuracy.To address this problem,in this study,we propose a new statistics-based algorithm,namely Particle Swarm Searching and Estimating(PSSE),which combines the fast searching way of Particle Swarm Optimization(PSO)with the accurately calculation of the Approximate Bayesian Computation(ABC).We first demonstrate that the new algorithm can recognize the parameters accurately and obtain the stable distributions of parameters,i.e.,consistent entropy information and less variances.Then,to further test the performance of the proposed approach,we compare the time complexity with PSO and ABC by using the observed data from three different nonlinear models on the same computational environment.Numerical results show that our approach exhibits obvious superiority in parameter identification of Biochemical reaction systems. |
