Discovery of Genes Implied in Cancer by Genetic Algorithms and Association Rules Alejandro S´ anchez Medina 1 , Alberto Gil Pichardo 1 , Jose Manuel Garc´ ıa-Heredia 2 , and Mar´ ıa Mart´ ınez-Ballesteros 3(B ) 1 University of Sevilla, Seville, Spain alesanmed@alum.us.es , albgilpic@alum.us.es 2 Department of Vegetal Biochemistry and Molecular Biology, University of Seville, Seville, Spain jmgheredia@us.es 3 Department of Computer Science, University of Seville, Seville, Spain mariamartinez@us.es Abstract. This work proposes a methodology to identify genes highly related with cancer. In particular, a multi-objective evolutionary algo- rithm named CANGAR is applied to obtain quantitative association rules. This kind of rules are used to identify dependencies between genes and their expression levels. Hierarchical cluster analysis, fold-change and review of the literature have been considered to validate the relevance of the results obtained. The results show that the reported genes are consistent with prior knowledge and able to characterize cancer colon patients. Keywords: Data mining · Association rules · Gene expression · Cancer 1 Introduction The word cancer refers to a set of different complex diseases, characterized by an uncontrolled and pathogenic growth of cells as a tumor, driving to death without medical treatment. Due to the increase of life expectancy, cancer can be considered as the 21 st century’s disease. In fact, around one third of the total population will develop cancer during their lifetime [1]. This high percentage shows the importance of the development of adequate techniques for diagnosis and treatment of the different cancers. The origin of cancer is mainly genetic, due to the accumulation of mutations in cells during lifetime. This enables them to become independent of the extra- cellular matrix, losing adherent properties and allowing them to metastasize to other tissues. During the period 1989–2008, around 15 % of all deaths due to malignant tumours were caused by colorectal cancer [2]. Data Mining has been used in cancer context for the diagnosis and prognosis of breast cancer [3]. This technique can help to detect it earlier reducing the mortality risk in patients. Indeed, data mining has been used in breast cancer