Coevolutionary Method for Gene Selection and Parameter Optimization in Microarray Data Analysis

摘要

This paper presents a coevolutionary algorithm based personalized modeling (cEAP) for gene selection and parameter optimization for microarray data analysis. The classification of different tumor types is a main application in microarray data analysis and of great importance in cancer diagnosis and drug discovery. However, the construction of an effective classifier involves gene selection and parameter optimization, which poses a big challenge to bioinformatics research. We have explored cEAP algorithm on four benchmark microarray datasets for gene selection and parameter optimization. The experimental results have shown that cEAP is an efficient method for co-evolving complex optimization problems in microarray data analysis.