Selecting informative genes using a multiobjective evolutionary algorithm

Recent advances in biotechnology offer the ability to measure the levels of expression of thousands of genes in parallel. Analysis of such data can provide understanding and insight into gene function and regulatory mechanisms, and open a new opportunity to tissue classification. Since the first work of Golub et al. (1999, 2000) in cancer classification based on gene expression profile rather than on morphological appearance of the tumor, there have been several endeavors in this direction. However, these tasks are made more difficult due to the noisy nature of array data and the overwhelming number of genes. In this paper, we propose a solution to the problem of gene selection using a multiobjective evolutionary algorithm (MOEA). Results from experiments with benchmarking data sets are also given