Biclustering of Gene Expression Data Using EDA-GA Hybrid

Biclustering of gene expression data is an important technology for biologists. The biclustering problem is proven to be NP-hard. Several biclustering methods have been proposed to analyze the gene expression data including genetic algorithms (GAs). However, genetic algorithms usually converge slowly when they are used to solve the largest biclustering problem. In this paper, we present a new method, EDA-GA hybrid, to analyze the gene expression data. After testing on simulated data, we find the hybrid algorithm not only can converge quickly, but also can obtain the global solution.