Machine Learning for Drug-Target Interaction Prediction.

Area of Science:

• Computational chemistry
• Bioinformatics
• Machine learning in drug discovery

Background:

• Identifying drug-target interactions (DTIs) is crucial for efficient drug discovery.
• In vitro experiments for DTI identification are costly and time-consuming.
• Computational prediction methods offer a high-efficiency alternative.

Purpose of the Study:

• To provide a comprehensive overview of machine learning approaches for DTI prediction.
• To summarize frequently used databases in drug discovery.
• To discuss recent advancements, advantages, limitations, and future outlook of machine learning in DTI prediction.

Main Methods:

• Hierarchical classification of machine learning methods for DTI prediction.
• Review of representative and state-of-the-art DTI prediction algorithms.
• Comparison of the strengths and weaknesses of different machine learning categories.

Main Results:

• Machine learning offers promising strategies for efficient DTI prediction.
• Various databases and machine learning methods are available for DTI prediction.
• Recent advancements show significant progress in computational DTI prediction.

Conclusions:

• Machine learning-based DTI prediction is a vital and evolving field in drug discovery.
• This review serves as a reference and tutorial for researchers in machine learning-based DTI prediction.
• Addressing current challenges and exploring future outlooks will further enhance computational drug discovery efforts.