Clustering In The Gene Chip Study

Gene chips are essential tools for genetic research. By analyzing the gene chip data, we can get access to hidden biological information. Clustering is an important means to analyze genetic data. This paper focuses on the cluster analysis of gene chip data, including partitioned clustering and pattern-based approach.This paper solves two major problems of partitioned clustering. First of all, a possibilistic fuzzy clustering was introduced to analyze gene chip data, but it has the defect of the same clustering center. To avoid this defect, we added centers excluded items to the objective function to carry out effective control. Second, the existing validity indexes under a fuzzy clustering environment are not workable in possibilistic clustering models. This paper had the existing validity index redefined by replacing possibilistic c-membership with modified possibilistic c-membership, so that it was workable under a possibilistic clustering environment.The exploration of the theoretical study on pattern-based clustering model was done in this paper. We have a universal definition of pattern and rule function. A study of clustering operation principle is processed. All of the study provides a theoretical basis for multi-pattern models.