Computational　analysis　is　essential　to　transform　the　masses　of　microarray　data　into　a　mechanistic　understanding　of　cancer.　Here　we　presented　a　method　for　finding　cancer　gene　functional　modules　from　microarray　data　and　applied　it　to　colon　cancer.　Comparison　colon　cancer　gene　modules　with　normal　gene　modules　show　that　these　cancer　genes　involved　in　the　biological　processes　including:　regulation　of　cell　cycle,　cell　proliferation,　signal　transduction,　and　anti-apoptosis.　Functional　changes　were　accompanied　with　module　changes　The　effectiveness　and　feasibility　of　the　method　is　proved　by　the　results.