JMIR Publications

ABSTRACT

Construction of most knowledge graphs, including those COVID-19-related, are based upon the co-occurring biomedical entities retrieved from recent literature. However, the applications drawn from these graphs (e.g., association predictions amongst genes, drugs, and diseases) have a high probability of false-positive predictions as the co-occurrences in literature do not always mean a true biomedical association between two entities. Data quality plays an important role in training deep neural network models, however, most of the current works in this area were focused on improving a model’s performance with the assumption that the pre-processed data are clean. Here, we studied how to remove noise from raw knowledge graphs with limited labeled information. Two novel Generative Adversarial Network models, NetGAN and CELL, applied to both the synthetic dataset and real dataset for edge classification (i.e., link prediction) leveraging unlabeled link information. The performance of link prediction, especially in the extreme case of training data versus test data at a ratio of 1:9, demonstrated that the promised method still achieved favorable results (AUCROC > 0.8 for synthetic and 0.7 for real dataset) despite the limited amount of testing data available. Our preliminary findings showed the proposed method achieved promising results for removing noise in data preprocessing of the biomedical knowledge graph, and potentially improved the performance of downstream applications by providing cleaner data.