iDrug

iDrug integrates drug repositioning and drug-target prediction by applying cross-network embedding to heterogeneous drug–target–disease networks to predict drug-disease and drug-target interactions.

Key Features:

• Cross-Network Embedding: Uses cross-network embedding to learn lower-dimensional feature spaces for drugs, targets, and diseases within a heterogeneous network.
• Knowledge Transfer: Facilitates knowledge transfer between disease and target domains by leveraging a shared drug feature space.
• Joint Consideration: Jointly models drug–target interactions and their modulation of biological pathways to integrate drug repositioning and drug-target prediction.
• Superior Performance: Empirical evaluations on real-world datasets demonstrated performance superior to several state-of-the-art approaches for both learning tasks.

Scientific Applications:

• Drug Repositioning: Predicts novel drug-disease associations to identify new therapeutic uses for existing drugs.
• Drug-Target Prediction: Identifies potential drug-target interactions by embedding drugs and targets into a shared feature space.
• Disease Mechanism Analysis: Relates drug-target interactions to modulation of biological pathways and disease treatment outcomes.

Methodology:

Cross-network embedding to learn lower-dimensional representations for drugs, targets, and diseases in a heterogeneous network; knowledge transfer via a shared drug feature space; joint modeling of drug–target interactions and their effects on biological pathways; empirical evaluation on real-world datasets.