This research aims at comparing the performance of two approaches in predicting cancer genes, i.e oncogenes (OCG), tumor suppressor genes (TSG) and cancer-related genes (others), for three different types of cancers, namely, liver cancer, esophaguae cancer and breast cancer. Gene expression data are downloaded from EBI ArrayExpress. The student t-test approach and the Significance Analysis of Microarrays (SAM) are employed to identify cancer genes. It is found that t-test performs better for esophagaue caner and breast cancer, while SAM performs better for liver cancer. The web site of studying results is available at http://ppi.bioinfo.asia.edu.tw/P_Cancer_T_SAM/
