Graph Embedding on Biomedical Networks: Methods, Applications, and Evaluations

Motivation: Graph embedding learning which aims to automatically learn low-dimensional node representations has drawn increasing attention in recent years. To date, most recent graph embedding methods are mainly evaluated on social and information networks and have yet to be comprehensively studied on biomedical networks under systematic experiments and analyses. On the other hand, for a variety of biomedical network analysis tasks, traditional techniques such as matrix factorization (which can be seen as one type of graph embedding methods) have shown promising results, and hence there is a need to systematically evaluate more recent graph embedding methods (e.g., random walk-based and neural network-based) in terms of their usability and potential to further the state-of-the-art. Results: We conduct a systematic comparison of existing graph embedding methods on three important biomedical link prediction tasks: drug-disease association (DDA) prediction, drug-drug interaction (DDI)prediction, protein-protein interaction (PPI) prediction, and one node classification task, i.e., classifying the semantic types of medical terms (nodes). Our experimental results demonstrate that the recent graph embedding methods are generally more effective than traditional embedding methods. Besides, compared with two state-of-the-art methods for DDAs and DDIs predictions, graph embedding methods without using any biological features achieve very competitive performance. Moreover, we summarize the experience we have learned and provide guidelines for properly selecting graph embedding methods and setting their hyper-parameters. Availability: We develop an easy-to-use Python package with detailed instructions, BioNEV, available at:https://github.com/xiangyue9607/BioNEV, including all source code and datasets, to facilitate studying various graph embedding methods on biomedical tasks