The promise of functional genomics is to find the genes responsible for phenotypes by carefully analyzing patterns of expression. Calculation of the fold-differences in gene expression between two diseases has been used to find those genes whose expression successfully differentiates the diseases. However, generating a list of independent genes with differential expression in two diseases does not necessarily help determine how regulatory differences contribute to disease etiology. Genes with the highest differences in expression levels may represent dissimilarities at levels far downstream, compared with subtle changes or no changes at all in expression levels in the central controlling genes. Focusing on differences in models of genetic regulatory networks in acute lymphocytic leukemia and acute myelogenous leukemia, rather than on differences in expression levels, yields a short list of well-known oncogenes, genes known to be involved in leukemogenesis, genes near known chromosomal breakpoints associated with leukemia and genes with unknown but possibly related functions.