An Effective Gene Selection Method for Cancer Subtype Classification Based on Predatory Search Genetic Algorithm and Support Vector Machine

The classification of different cancer subtypes and feature subset selection is of great importance in cancer diagnosis and has recently received a great deal of attention in the field of bioinformatics. Rich Gene information can aid in the classification and recognition of the cancer subtype. Meanwhile, the high dimensions of gene also cause redundancy in information and it should be reduced. This paper proposed an effective feature gene selection method based on predatory search genetic algorithm (PSGA), which combined predatory search with genetic algorithm to solve the problems of the existing gene selection methods. PSGA was proposed in order to choose the higher correlation genes to the classification task, and support vector machine (SVM) classification method was implemented to classify different cancer types. Experimental results show that the proposed method can achieve higher classification accuracy by using PSGA and SVM.