Ovarian cancer is a morphologically and biologically heterogeneous disease. The identification of type-specific protein markers for ovarian cancer would provide the basis for more tailored treatments, as well as clues for understanding the molecular mechanisms governing cancer progression. In the present study, we used a novel approach to classify 24 ovarian cancer tissue samples based on the proteomic pattern of each sample. The method involved fractionation according to pI using chromatofocusing with analytical columns in the first dimension followed by separation of the proteins in each pI fraction using nonporous RP HPLC, which was coupled to an ESI-TOF mass analyzer for molecular weight (MW) analysis. A 2-D mass map of the protein content of each type of ovarian cancer tissue samples based upon pI versus intact protein MW was generated. Using this method, the clear cell and serous ovarian carcinoma samples were histologically distinguished by principal component analysis and clustering analysis based on their protein expression profiles and subtype-specific biomarker candidates of ovarian cancers were identified, which could be further investigated for future clinical study.