Ewing's sarcomas contain specific chromosomal translocations that fuse EWS to ETS family members, including FLI, ERG, FEV, ETV1 and ETV4. Prior work has suggested that functional differences exist between some of these EWS-ETS fusions. However, as the cell of origin of Ewing's sarcoma is unknown, this prior work was conducted in NIH3T3 cells, which have not been validated as an appropriate model for the study of EWS-ETS fusions. To determine if NIH3T3 cells are a good model for Ewing's sarcoma, we introduced all five EWS-ETS fusions into these cells, and analyzed their phenotypes and gene expression patterns. EWS-FLI, EWS-ERG, and EWS-FEV caused NIH3T3 cells to exhibit anchorage independent growth whereas EWS-ETV1 and EWS-ETV4 did not. In contrast, all the EWS-ETS fusions induced tumor formation in a xenograft model. We defined the core transcriptional profile of the EWS-ETS fusions using cDNA microarrays, and compared these to data derived from patient-derived Ewing's sarcoma cell lines. The NIH3T3 model did not recapitulate the gene expression pattern of bona fide Ewing's sarcoma. Based on these results, we conclude that while there may be functional differences between the various EWS-ETS fusions, the NIH3T3 cell model is inadequate to study the gene expression pattern induced by EWS-ETS proteins in Ewing's sarcoma. Thus, data derived from the NIH3T3 model system needs to be appropriately validated before they can be accepted as relevant to the human disease.