+1 Recommend
    • Review: found
    Is Open Access

    Review of 'Explainable drug repurposing via path-based knowledge graph completion'

    Explainable drug repurposing via path-based knowledge graph completionCrossref
    Interesting tool for mining knowledgegraphs for new information that could be explained through them
    Average rating:
        Rated 4.5 of 5.
    Level of importance:
        Rated 5 of 5.
    Level of validity:
        Rated 5 of 5.
    Level of completeness:
        Rated 4 of 5.
    Level of comprehensibility:
        Rated 4 of 5.
    Competing interests:

    Reviewed article

    • Record: found
    • Abstract: found
    • Article: found
    Is Open Access

    Explainable drug repurposing via path-based knowledge graph completion

    Drug repurposing aims to find new therapeutic applications for existing drugs in the pharmaceutical market, leading to significant savings in time and cost. The use of artificial intelligence and knowledge graphs to propose repurposing candidates facilitates the process, as large amounts of data can be processed. However, it is important to pay attention to the explainability needed to validate the predictions. We propose a general architecture to understand several explainable methods for graph completion based on knowledge graphs and design our own architecture for drug repurposing. We present XG4Repo (eXplainable Graphs for Repurposing), a framework that takes advantage of the connectivity of any biomedical knowledge graph to link compounds to the diseases they can treat. Our method allows methapaths of different types and lengths, which are automatically generated and optimised based on data. XG4Repo focusses on providing meaningful explanations to the predictions, which are based on paths from compounds to diseases. These paths include nodes such as genes, pathways, side effects, or anatomies, so they provide information about the targets and other characteristics of the biomedical mechanism that link compounds and diseases. Paths make predictions interpretable for experts who can validate them and use them in further research on drug repurposing. We also describe three use cases where we analyse new uses for Epirubicin, Paclitaxel, and Predinisone and present the paths that support the predictions.

      Review information

      This work has been published open access under Creative Commons Attribution License CC BY 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Conditions, terms of use and publishing policy can be found at www.scienceopen.com.

      Bioinformatics & Computational biology
      Interpretability,Hetionet,Heterogeneous Knowledge Graphs,Rule-based link prediction,Drug Repurposing,Knowledge Graph Completion

      Review text

      The article titled "Explainable drug repurposing via path-based knowledge graph completion" by Ana Jimenez et al. is focused on using artificial intelligence and knowledge graphs for drug repurposing. The study aims to find new therapeutic applications for existing drugs using AI and knowledge graphs. The methodology makes use of paths within the knowledge graph for drug repurposing. This approach not only predicts which diseases can be treated by a drug but also provides a biological rationale for the predictions. The method delves into how paths, which are sequences of nodes and relationships in the graph, and metapaths, a specific sequence of node and relation types, are used in this process. The paper outlines different strategies for generating these paths, highlighting their importance in making the drug repurposing predictions interpretable and meaningful for further research.

      Here is a review based on the provided criteria:

      • The model's focus on paths from compounds to diseases might limit its ability to uncover new, less obvious drug-disease associations, potentially leading to a focus on already known cases. The inclusion of different node types such as genes, pathways, side effects, or anatomies might mitigate this to some extent, but the primary focus remains on compounds and diseases.
      • The model is trained to give high scores to diseases known to be treated with the compound. This training approach could potentially lead to overfitting, where the model might excel in identifying known cases but struggle with novel or less obvious associations. A discussion about this issue is suggested for inclusion in the main text.
      • The selection of examples using Epirubicin, Paclitaxel, and Prednisone could be justified in the main text.
      • The choice of knowledge graph used for training XG4Repo and the effect of changing this knowledge graph are not discussed, which is significant for its adaptability to different datasets and scenarios.

      The work is original, providing new insights into explainable AI for drug repurposing. Its significance lies in offering a framework (XG4Repo) that combines prediction accuracy and interpretability, potentially impacting how researchers approach drug repurposing.


      Comment on this review