Schwann cells constitute the major cell population in neurofibromas, tumors that disfigure patients with the common inherited disease neurofibromatosis type 1. It is not known how abnormalities in neurofibroma Schwann cells contribute to tumor formation. We developed phenotypic assays that distinguish normal Schwann cells from Schwann cells (mouse and human) that lack the NF1 tumor suppressor protein, a GTPase-activating protein for Ras proteins. To define genes that may underlie these phenotypic changes we hybridized cDNA from wild-type and Nf1 mutant mouse Schwann cells to microarrays from Incyte Genomics and compared gene expression levels. We carried out four arrays, each in duplicate, comparing cells of various phenotypes (Nf1+/−, Nf−/−, Nf1−/−TXF [a proliferative Schwann cell population] and Nf1−/−TXF treated with a farnesyl transferase inhibitor [a drug class in clinical trials in NF1] to wild-type Schwann cells. Thirteen changes of greater than fourfold were identified in Nf1−/− Schwann cells. Larger numbers of genes were differentially expressed in Nf1−/−TXF Schwann cells; some were sensitive to farnesyl transferase inhibitor. We report on the reproducibility of the data and confirm the RNA and protein levels for some identified genes. We have also begun an effort to share array data among laboratories studying neurofibromatosis as part of an international consortium. By applying these methods to NF1 model systems we hope to learn how neurofibromin serves as a tumor suppressor and to identify potential biomarkers and therapeutic targets for improved management of neurofibromatosis.