Research On Gene Expression Data Clustering Algorithm Based On Particle Swarm Optimization

Abstract:With the development of MicroArray technology, more and more gene expression datasets are being obtained. So, how the useful information can be drawn from the gene expression datasets becomes an important issue in the Bioinformatic research field. Those genes with similar functions usually share similar expression patterns. The unknown genes’function can be forecasted by analyzing genes with similar expression pattern. It is necessary to adopt proper analyzing algorithm to deal with complex gene expression. In this thesis, the clustering technique is used to deal with the gene expression data, several representative clustering algorithms are introduced such as K-means algorithms, self-organizing feather competitive network, and so on.In this paper, the PSO algorithm is applied in the fields of gene expression data cluster analysis on the basis of in-depth analysis of the basic theory of the PSO algorithm, and some improvements are proposed to the algorithm. In order to avoid premature convergence of Particle Swarm Optimization (PSO), a new PSO algorithm based on adaptive disturbance (ADPSO) is proposed to help trapped particles escape from local minima. Then the improved algorithm is combined with the K-means algorithm, the gene expression data clustering algorithm based on ADPSO-KM is proposed. The algorithm effectively combines the advantages of the K-means algorithm and PSO algorithm.For investigating the convergence performance and the optimized ability of the ADPSO-KM, verifying the superiority of the algorithm, four datasets are choosed to conduct the numerical simulations of K-means algorithm, PSO algorithm and ADPSO-KM algorithm. The results show that the ADPSO-KM is able to find better solutions on most test functions, and it is a global convergent optimization algorithm with high convergent precision and fast convergent speed.