Knowledge Graphs for drug repurposing: a review of databases and methods

Area of Science:

• Pharmacology
• Biotechnology
• Computer Science

Background:

• Drug repurposing offers an efficient strategy for identifying novel therapeutic agents.
• Knowledge Graphs (KGs) are increasingly utilized for discovering potential drug candidates.

Purpose of the Study:

• To review prominent Knowledge Graphs (KGs) and their role in drug repurposing.
• To explore artificial intelligence (AI) techniques that enhance drug repurposing efficiency and precision.
• To emphasize the importance of explainability and validation in AI-driven drug repurposing.

Main Methods:

• Comprehensive review of existing Knowledge Graphs (KGs), detailing their structure and data sources.
• Exploration of various artificial intelligence (AI) techniques applied to drug repurposing.
• Discussion of explainability methods and prediction validation strategies.

Main Results:

• Knowledge Graphs (KGs) provide a robust framework for identifying drug repurposing candidates.
• AI techniques significantly accelerate and improve the accuracy of drug repurposing predictions.
• Explainability methods enhance the trustworthiness and transparency of AI-driven drug discovery.

Conclusions:

• The integration of KGs and AI offers a powerful approach to drug repurposing.
• Explainable AI (XAI) is crucial for validating and trusting AI-generated drug repurposing predictions.
• Further research into validation techniques will ensure reliable and understandable drug discovery outcomes.