Molecular disease mechanisms typically constitute abnormalities in the regulation of genes producing many kinds of alterations in the expression levels. To identify disease-altered genes better, we have developed an approach that searches for the genes which present a significant rank alteration in the rank of their expression profiles, by comparing an altered rank with another gene. The approach provides groups of genes and assigns a statistical measure of significance to each pair of genes selected. The method is evaluated using two experimental sets of microarrays. We investigated the possibility of inferring the rank reverse pairs and its possibility in diagnostic application when there is no prior knowledge of the genes belonging to disease, nor about the structure of gene regulatory network. We also show that rank reverse is a very powerful technique to diagnostic disease in the practical application and could be used prior to traditional biomarker discovery.