PREDICTING GENE DISEASE ASSOCIATIONS WITH KNOWLEDGE GRAPH EMBEDDINGS FOR DISEASES WITH CURTAILED INFORMATION

Abstract

Knowledge graph embeddings (KGE) are a powerful technique used in the biological domain to represent biological knowledge in a low dimensional space. However, a deep understanding of these methods is still missing, and in particular the limitations for diseases with reduced information on gene-disease associations. In this contribution, we built a knowledge graph (KG) by integrating heterogeneous biomedical data and generated KGEs by implementing state-of-the-art methods, and two novel algorithms: DLemb and BioKG2Vec. Extensive testing of the embeddings with unsupervised clustering and supervised methods showed that our novel approaches outperform existing algorithms in both scenarios. Our results indicate that data preprocessing and integration influence the quality of the predictions and that the embeddings efficiently encodes biological information when compared to a null model. Finally, we employed KGE to predict genes associated with Intervertebral disc degeneration (IDD) and showed that functions relevant to the disease are enriched in the genes prioritized from the model GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=110 SRC="FIGDIR/small/575314v1_ufig1.gif" ALT="Figure 1"> View larger version (21K): org.highwire.dtl.DTLVardef@1ba98b7org.highwire.dtl.DTLVardef@1801d6borg.highwire.dtl.DTLVardef@b87a7org.highwire.dtl.DTLVardef@f70cf8_HPS_FORMAT_FIGEXP M_FIG C_FIG