Ovarian cancer has the worst prognosis of any gynecological malignancy, and generally presents with metastasis at advanced stages. Copy number variation (CNV) frequently contributes to the alteration of oncogenic drivers. In this study, we sought to identify genetic targets in heterogeneous clones from human ovarian cancers cells. We used array-based technology to systematically assess all the genes with CNVs in cell models clonally expanded from A2780 and SKOV3 ovarian cancer cell lines with distinct highly and minimally invasive/migratory capacities. We found that copy number alterations differed between matched highly and minimally invasive/migratory subclones, differentially affecting specific functional processes including immune response processes, DNA damage repair, cell cycle and cell proliferation. We also identified seven genes as strong candidates, including DDB1, ERCC1, ERCC2, PRPF19, BCAT1, CDKN1B and MARK4, by integrating the above data with gene expression and clinical outcome data. Thus, by determining the molecular signatures of heterogeneous invasive/migratory ovarian cancer cells, we identified genes that could be specifically targeted for the treatment and prognosis of advanced ovarian cancers.