摘要

In order to propose a novel biclustering algorithm and to cluster gene expression data, this paper improved the residue function of the biclustering algorithm. The improved function can be more appropriate used for stochastic heuristic searching algorithm. The parallel genetic algorithm was firstly introduced to the of the biclustering optimization algorithm. Hence this method can prevent local convergence in the optimal algorithm in a great extent, and make the probability of approaching the global convergence bigger. The algorithm used the Yeast Saccharomyces cerevisiae cell cycle gene expression profile from the data Spellman used in his own experiment in Stanford database to bicluster. And we compared our algorithm with traditional genetic algorithms in biclustering. The efficiency of this algorithm was proved.

  • 出版日期2012

全文