In silico methods for drug-target interaction prediction.

Area of Science:

• Computational chemistry
• Bioinformatics
• Drug discovery

Background:

• Drug-target interaction (DTI) prediction is vital for drug discovery.
• In silico methods offer a cost-effective and efficient alternative to traditional drug development.
• The increasing volume of biological data necessitates advanced computational approaches for DTI prediction.

Purpose of the Study:

• To review current in silico DTI prediction approaches.
• To identify key factors influencing DTI prediction accuracy.
• To propose strategies for overcoming challenges in DTI prediction.

Main Methods:

• Analysis of existing literature on DTI prediction.
• Identification of four major factors impacting DTI predictions.
• Evaluation of data, feature engineering, and experimental setup strategies.
• Exploration of refining "guilt-by-association" methods.
• Integration of emerging technologies like large language models and AlphaFold.

Main Results:

• Four key factors influencing DTI prediction were identified.
• Strategies for managing data sparsity were proposed, including refining "guilt-by-association".
• The potential of large language models and AlphaFold for feature engineering in DTI prediction was highlighted.
• Persistent challenges in DTI prediction were discussed.

Conclusions:

• Refining established methods and integrating novel technologies are crucial for advancing DTI prediction.
• This review provides guidance for future research in computational drug discovery.
• Improved DTI prediction can accelerate the drug development process.