DRML-Ensemble: drug repurposing method based on feature construction of multi-layer ensemble.

Area of Science:

• Computational biology
• Pharmacology
• Bioinformatics

Background:

• Drug repurposing accelerates drug development by identifying new uses for existing drugs, reducing costs.
• Proteins are crucial as drug targets or products of disease-related genes in drug repurposing.
• Extracting high-quality protein features from heterogeneous graphs remains a challenge.

Purpose of the Study:

• To propose a novel drug repurposing model, DRML-Ensemble, addressing challenges in protein feature extraction from heterogeneous graphs.
• To enhance drug repurposing accuracy by leveraging multi-layer feature construction.

Main Methods:

• Developed DRML-Ensemble utilizing multiple layers of heterogeneous graph feature construction (HGFC).
• HGFC layers extract protein features by integrating drug, disease, and protein relationships.
• A multi-layer ensemble prediction module is incorporated to optimize performance and mitigate overfitting.

Main Results:

• DRML-Ensemble achieved an Area Under the Precision-Recall curve (AUPR) of 0.93 and an Area Under the Receiver Operating Characteristic curve (AUROC) of 0.92.
• The model's performance surpassed existing state-of-the-art drug repurposing methods.
• Demonstrated significant potential for drug repurposing in Alzheimer's disease.

Conclusions:

• DRML-Ensemble effectively extracts protein features from heterogeneous graphs for improved drug repurposing.
• The proposed model offers a promising computational approach for identifying novel drug candidates.
• The study highlights the utility of DRML-Ensemble in tackling complex diseases like Alzheimer's.