DTKGIN: Predicting drug-target interactions based on knowledge graph and intent graph.

Area of Science:

• Computational biology
• Drug discovery
• Bioinformatics

Background:

• Predicting drug-target interactions (DTIs) is vital for efficient drug discovery and development.
• Computational methods reduce the high cost and risks associated with experimental approaches.
• Existing DTI prediction methods face challenges with data sparsity and the cold-start problem.

Purpose of the Study:

• To develop an advanced computational model for predicting drug-target interactions.
• To address limitations of existing methods, specifically data sparsity and the cold-start problem.
• To enhance the accuracy and efficiency of drug discovery pipelines.

Main Methods:

• A novel model, DTKGIN, was developed using a knowledge graph and an intent graph.
• DTKGIN captures biological environmental information by mining drug and target relations within a knowledge graph.
• It considers fine-grained drug-target interactions via an intent graph, learning representations for drugs and targets.

Main Results:

• DTKGIN demonstrated superior performance compared to state-of-the-art methods in 10-fold cross-validation.
• The model showed particular effectiveness in cold-start experimental settings.
• Case studies confirmed DTKGIN's capability in identifying potential drug-target interactions.

Conclusions:

• DTKGIN offers a robust and effective approach for predicting drug-target interactions.
• The model's ability to handle sparsity and cold-start problems advances computational drug discovery.
• DTKGIN provides a valuable tool for researchers aiming to accelerate the identification of novel therapeutic targets.