Drug Repositioning using Consilience of Knowledge Graph Completion Methods
Tu, R.; Sinha, M.; Gonzalez-Cavazos, A. C.; Hu, E.; Dhuliawala, S. Z.; McCallum, A.; Su, A. I.
Show abstract
While link prediction methods in knowledge graphs have been increasingly utilized to locate potential associations between compounds and diseases, they suffer from lack of sufficient evidence to explain why a drug and a disease may be indicated. This is especially true for knowledge graph embedding (KGE) based methods where a drug-disease indication is linked only by information gleaned from a vector representation. Complementary pathwalking algorithms can increase the confidence of drug repurposing candidates by traversing a knowledge graph. However, these methods heavily weigh the relatedness of drugs, through their targets, pharmacology or shared diseases. Furthermore, these methods can rely on arbitrarily extracted paths as evidence of a compound to disease indication and lack the ability to make predictions on rare diseases. In this paper, we evaluate seven link prediction methods on a vast biomedical knowledge graph for drug repurposing. We follow the principle of consilience, and combine the reasoning paths and predictions provided by path-based reasoning approaches with those of KGE methods to identify putative drug repurposing indications. Finally, we highlight the utility of our approach through a potential repurposing indication.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- A Survey and Systematic Assessment of Computational Methods for Drug Response Prediction 97%
- Interpretable and Generalizable Attention-Based Model for Predicting Drug-Target Interaction Using 3D Structure of Protein Binding Sites: SARS-CoV-2 Case Study and in-Lab Validation 96%
- A geometric deep learning framework for drug repositioning over heterogeneous information networks 95%
Similar papers in this journal
- A Graph-Attention-Based Deep Learning Network for Predicting Biotech-Small-Molecule Drug Interactions 97%
- Mining drug-target interactions from biomedical literature using chemical and gene descriptions-based ensemble transformer model. 96%
- Prompt-to-Pill: Multi-Agent Drug Discovery and Clinical Simulation Pipeline 96%
Similar papers in this journal
- Network-based estimation of therapeutic efficacy and adverse reaction potential for prioritisation of anti-cancer drug combinations 96%
- TCMM: A Unified Database for Traditional Chinese Medicine Modernization and Therapeutic Innovations 96%
- G-PLIP: Knowledge graph neural network for structure-free protein-ligand bioactivity prediction 95%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.