Several, acquired, non-random chromosomal deletions have been characterized in acute myelogenous leukemia (AML). While the deletion limits vary among patients, there are consistent regions of overlap among the deleted segments between patients. Furthermore, chromosomal deletions are achieved frequently by unbalanced translocations between two and more chromosomes resulting loss of candidate leukemia suppressor loci from the affected chromosomes. Most deletions occurring as sole anomalies are associated with good-intermediate clinical outcome, but complex cytogenetic anomalies signify an aggressive clinical course. Thanks to the exciting development in microarray, siRNA technologies, a number of candidate AML suppressor genes localizing to the critical regions of overlap within the deletions have been identified recently. Most of the candidate genes do not function by the classical "two hits," namely loss of an allele unmasking inactivating mutations in the remaining allele. Gene dosage, epigenetic silencing, and uniparental disomy appear to be common mechanisms of gene inactivation in AML. While several of the newly discovered candidate genes lead to new pathways, a few of them affect previously known leukemogenic targets. Thus the investments made over the years on leukemia suppressor gene discovery are beginning to yield reasonable results at the present time. Future beholds promise for targeted therapy of these poorly characterized AMLs, as we uncover the mutations driving their clonal evolution.