Design and application of a knowledge network for automatic prioritization of drug mechanisms

Mayers, Michael; Tu, Roger; Steinecke, Dylan; Li, Tong Shu; Queralt-Rosiñach, Núria; Su, Andrew I.

doi:10.1093/bioinformatics/btac205

Cited by 15 publications

(10 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on this definition of correct matched paths, we finally find 472 unique drug-disease pairs of which each has at least one such correct matched path in its all possible 3-hop paths between drug and disease in our customized biomedical knowledge graph. For those 2,421 unique drug-disease pairs that are filtered out because of no such correct matched path, most of them are due to the missing edges in our customized biomedical knowledge graph which is consistent with the findings in the previous analysis [Mayers et al, 2022]. There are two reasons that could be used to explain these missing edges: 1) we filter out many low-quality edges from SemMedDB (described in Appx.…”

Section: Data Collection and Pre-processingsupporting

confidence: 77%

“…Since these MOA paths are extracted from the human free-text description, the length of these MOA paths is varying. Based on a previous report [Mayers et al, 2022], there are many edges/relations of these MOA paths that are missing in the existing biomedical databases and thus it is difficult to match the complete MOA paths to the KG-based paths in our customized biomedical knowledge graph. In addition, since the length of predicted paths generated by our model is fixed to 3, we consider those 3-hop KG-based paths of which all four nodes show up in the complete MOA paths are the correct matched paths.…”

Section: A32 Pre-processingmentioning

confidence: 99%

See 1 more Smart Citation

KGML-xDTD: A Knowledge Graph-based Machine Learning Framework for Drug Treatment Prediction and Mechanism Description

Liu

Zhou

et al. 2022

Preprint

View full text Add to dashboard Cite

Computational drug repurposing is a cost- and time-efficient method to identify new indications of approved or experimental drugs/compounds. It is especially critical for emerging and/or orphan diseases due to its cheaper investment and shorter research cycle compared with traditional wet-lab drug discovery approaches. However, the underlying mechanisms of action between repurposed drugs and their target diseases remain largely unknown, which is still an unsolved issue in existing repurposing methods. As such, computational drug repurposing has not been widely adopted in clinical settings. In this work, based on a massive biomedical knowledge graph, we propose a computational drug repurposing framework that not only predicts the treatment probabilities between drugs and diseases but also predicts the path-based, testable mechanisms of action (MOAs) as their biomedical explanations. Specifically, we utilize the GraphSAGE model in an unsupervised manner to integrate each entity's neighborhood information and employ a Random Forest model to predict the treatment probabilities between pairs of drugs and diseases. Moreover, we train an adversarial actor-critic reinforcement learning model to predict the potential MOA for explaining drug purposing. To encourage the model to find biologically reasonable paths, we utilize the curated molecular interactions of drugs and a PubMed-publication-based concept distance to extract potential drug MOA paths from the knowledge graph as "demonstration paths" to guide the model during the process of path-finding. Comprehensive experiments and case studies show that the proposed framework outperforms state-of-the-art baselines in both predictive performance of drug repurposing and explanatory performance of recapitulating human-curated DrugMechDB-based paths.

show abstract

Section: Data Collection and Pre-processingsupporting

confidence: 77%

Section: A32 Pre-processingmentioning

confidence: 99%

KGML-xDTD: A Knowledge Graph-based Machine Learning Framework for Drug Treatment Prediction and Mechanism Description

Liu

Zhou

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Mechanistic Repositioning Network with Indications (MIND)is a knowledge graph that distinguishes approved drug indications from semantically derived drug-disease relationships. Based on MechRepoNet, a knowledge graph that reflects important drug mechanism relationships identified from a curated biomedical drug mechanism dataset, MIND elevates pre-existing DrugCentral (a curated resource with regulatory approved drug indications) obtained treat edges as indication edges (25; 26). The treat edge represents a weaker link between a drug and disease compared to the indication edge as treat edges are not substantiated by regulatory approval.…”

Section: Methodsmentioning

confidence: 99%

“…al. expanded on this approach and incorporated rules based path exclusions and hyperparameter optimization schemes to improve path interpretability for drug repositioning (25).…”

Section: Methodsmentioning

confidence: 99%

Drug Repurposing using consilience of Knowledge Graph Completion methods

Sinha

González

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

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 prove 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 repositioning 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 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 large biomedical knowledge graph on a drug repositioning task. We follow the principle of consilience, and combine the reasoning paths and predictions provided by probabilistic case-based reasoning (probCBR) with those of a KGE method, TransE, to identify putative drug repositioning indications. Finally, we highlight the utility of our approach through two potential repurposing indications.

show abstract

“…KGs describe relationships between entities of different types, e.g., how a gene or gene product influences a disease, or how a metabolite is processed in a specific pathway. KGs have broad applications across biomedical research, including prediction of drug-drug interactions [8,9], evaluation of mechanisms of toxicity [10], prediction of unknown drug disease targets [11], or linking symptoms from electronic health records to better understand disease inference [12] . However, these applications have yet to be extended into the microbiome field [7].…”

Section: Introductionmentioning

confidence: 99%

Hypothesizing mechanistic links between microbes and disease using knowledge graphs

Santangelo,

Bada,

Hunter

et al. 2023

Preprint

View full text Add to dashboard Cite

Knowledge graphs have found broad biomedical applications, providing useful representations of complex knowledge. Although plentiful evidence exists linking the gut microbiome to disease, mechanistic understanding of those relationships remains generally elusive. Here we demonstrate the potential of knowledge graphs to hypothesize plausible mechanistic accounts of host-microbe interactions in disease. To do so, we constructed a knowledge graph of linked microbes, genes and metabolites called MGMLink. Using a semantically constrained shortest path search through the graph and a novel path prioritization methodology based on cosine similarity, we show that this knowledge supports inference of mechanistic hypotheses that explain observed relationships between microbes and disease phenotypes. We discuss specific applications of this methodology in inflammatory bowel disease and Parkinson’s disease. This approach enables mechanistic hypotheses surrounding the complex interactions between gut microbes and disease to be generated in a scalable and comprehensive manner.

show abstract

Design and application of a knowledge network for automatic prioritization of drug mechanisms

Cited by 15 publications

References 50 publications

KGML-xDTD: A Knowledge Graph-based Machine Learning Framework for Drug Treatment Prediction and Mechanism Description

KGML-xDTD: A Knowledge Graph-based Machine Learning Framework for Drug Treatment Prediction and Mechanism Description

Drug Repurposing using consilience of Knowledge Graph Completion methods

Hypothesizing mechanistic links between microbes and disease using knowledge graphs

Contact Info

Product

Resources

About