DrugMechDB: A Curated Database of Drug Mechanisms

Area of Science:

• Pharmacology
• Bioinformatics
• Computational Biology

Background:

• Computational drug repositioning accelerates drug development by identifying new uses for existing drugs.
• Biomedical knowledge graphs provide biological evidence for drug repositioning through reasoning chains.
• A lack of standardized databases for drug mechanisms hinders the training and evaluation of computational models.

Purpose of the Study:

• To introduce the Drug Mechanism Database (DrugMechDB), a manually curated resource for drug mechanisms.
• To provide a benchmark dataset for assessing computational drug repositioning methods.
• To serve as a valuable resource for training drug repositioning models.

Main Methods:

• Manual curation of drug mechanisms from diverse free-text resources.
• Representation of drug mechanisms as paths within a knowledge graph.
• Integration of data across 14 major biological scales.

Main Results:

• DrugMechDB contains 4,583 drug indications with 32,249 relationships.
• The database describes drug mechanisms as paths through a knowledge graph.
• The resource integrates information from authoritative free-text sources.

Conclusions:

• DrugMechDB addresses the need for a standardized drug mechanism database.
• The database can be used to train and evaluate computational drug repositioning models.
• DrugMechDB facilitates more effective and efficient drug repositioning strategies.