Integrated omics analysis based on transcriptome and DNA methylation data combined with machine learning methods is very promising for the diagnosis, prognosis, and classification of cancer. In this study, the DNA methylation and gene expression data of ovarian cancer (OC) were analyzed to identify abnormally methylated differentially expressed genes (DEGs), screen potential therapeutic agents for OC, and construct a risk model based on the abnormally methylated DEGs to predict patient prognosis. The gene expression and DNA methylation data of primary OC samples with tumor protein 53 (TP53) wild-type and normal samples were obtained from The Cancer Genome Atlas (TCGA) database. DEGs with aberrant methylation were analyzed by screening the intersection between DEGs and differentially methylated genes (DMGs). We attempted to search for potential drugs targeting DEGs with aberrant methylation by employing a network medicine framework. A gene signature based on the DEGs with aberrant methylation was constructed by regularized least absolute shrinkage and selection operator (LASSO) regression analysis. A total of 440 aberrant methylated DEGs were screened. Based on their gene expression profiles and methylation data from different regions, the results of both discriminative pattern recognition analysis and principal component analysis (PCA) showed a significant separation between tumor tissue and healthy ovarian tissue. In total, 126 potential therapeutic drugs were identified for OC by network-based proximity analysis. Five genes were identified in 440 aberrant methylated DEGs, which formed an aberrant methylated DEGs-driven gene signature. This signature could significantly distinguish the different overall survivals (OS) of OC patients and showed better predictive performance in both the training and validation sets. In this study, the DNA methylation and gene expression data of OC were analyzed to identify abnormally methylated DEGs and potential therapeutic drugs, and a gene signature based on five aberrant methylation DEGs was constructed, which could better predict the risk of death in patients.