Exploiting the transcriptional regulation mechanism by microarray and bioinformatics is an important method at genomic research field, and it is better than previous methods, by which we can easily analyze the gene expression regulative networks at genomic level. This study was designed to analyze the glioma gene expression profiles by clustering and bioinformatic retrieving to seek some tumor development relative genes that can be cloned for the purpose of function research in the future.Firstly, we analyzed the glioma gene expression profiles of 16,363 genes by clustering, and chose 368 genes which expression differences were greater than 2 folds and the variances of 2 dots were smaller than 0.33. We selected 2 groups of genes that were expressed similarly. One group (11 cases) was upregulated with the glioma development and another (6 cases) was downregulated with the glioma development. Secondly, we determined the genomic information about those 17 genes and exploited their association with the development of gliomas.Two groups of genes were expressed similarly during the glioma development, one group was upregulated with the development of gliomas, and another was downregulated. Bioinformatic analysis showed that 3 of those genes (X55987, M85085, and AB011097) might be important tumor relative genes.By bioinformatics and microarray technologies, we found 3 genes, X55987(EDN, eosinophil-derived neurotoxin), AB011097 (ARTS-1, TNF receptor shedding regulator),and M85085(CStF, cleavage stimulation factor), which might be potential key tumor development relative genes that can be developed for therapy targets in the future.
