Objective: To study the gene expression of highmetastatic human ovarian carcinoma cell line(HO-8910PM) and to screen for novel metastasis- associated genes by cDNA microarray. Methods: The cDNA wasretro-transcribed from equal quantity mRNA derived from tissues of highly metastatic ovarian carcinoma cell line and normal ovarian, and was labeled with Cy5 and Cy3fluorescence as probes. The mixed probes were hybridized with BioDoor 4096 double dot human whole gene chip. The chip was scanned by scanArray 3000 laser scanner. Theacquired image was analyzed by ImaGene 3.0 software.Results: By applying the cDNA microarray we found: Atotal of 323 genes whose expression level were 3 timeshigher or lower in HO-8910PM cell than normal ovarianepithelium cell were screened out, with 71 higher and 252lower respectively. Among these 10 were new genes. 67genes showed expression difference bigger than 6 timesbetween HO-8910PM cell and normal ovarian epitheliumcell, among these genes 12 were higher, 55 lower, and twonew genes were found. Conclusion: cDNA microarraytechnique is effective in screening the differentiallyexpressed genes between human ovarian cancer cell line(HO-8910PM) and normal ovarian epithelium cell. Usingthe cDNA microarray to analyze of human ovarian cancer cell line gene expression profile difference will help the gene diagnosis, treatment and protection. Objective: To study the gene expression of high metaastatic human ovarian carcinoma cell line (HO-8910PM) and to screen for novel metastasis- associated genes by cDNA microarray. Methods: The cDNA wasretro-transcribed from equal quantity mRNA derived from tissues of highly metastatic ovarian The mixed cell lines and normal ovarian, and was labeled with Cy5 and Cy3fluorescence as probes. The mixed probes were hybridized with BioDoor 4096 double dot human whole gene chip. The chip was scanned by scanArray 3000 laser scanner. Theacquired image was analyzed by ImaGene 3.0 software. .Results: By applying the cDNA microarray we found: Atotal of 323 genes whose expression level were 3 timeshigher or lower in HO-8910PM cell than normal ovarianepithelium cells were screened out, with 71 higher and 252lower respectively. Among these 10 were new genes. 67genes showed expression difference bigger than 6 timesbetween HO-8910PM cell and normal ovarian epitheliumcell, among these genes 12 were higher, 55 lo Wer, and twonew genes were found. Conclusion: cDNA microarraytechnique is effective in screening the differentiallyexpressed genes between human ovarian cancer cell line(HO-8910PM) and normal ovarian epithelium cell. Usingthe cDNA microarray to analyze of human ovarian cancer cell line gene expression profile Difference will help the gene diagnosis, treatment and protection.