Comprehensive Review on Drug-target Interaction Prediction - Latest Developments and Overview

Area of Science:

• Pharmacology
• Computational Biology
• Drug Discovery

Background:

• Drug-target interactions (DTIs) are crucial for modulating biological functions and form the basis of drug development.
• Experimental methods for identifying DTIs are time-consuming and expensive, necessitating computational approaches.
• Accurate DTI prediction can significantly enhance the efficiency and reduce the costs associated with drug discovery.

Purpose of the Study:

• To provide a comprehensive overview of drug-target interactions as a critical initial step in drug discovery.
• To explore the application and effectiveness of machine learning methods in predicting DTIs.
• To review relevant literature and databases used in the field of DTI prediction.

Main Methods:

• Review of existing literature and databases focused on drug-target interactions.
• Exploration of computational methods for DTI prediction, including docking simulations, ligand-based approaches, and machine learning techniques.
• Analysis of machine learning applications for predicting associations between drugs and their biological targets.

Main Results:

• Drug-target interaction prediction is a vital area within drug discovery, impacting efficiency and cost reduction.
• Machine learning methods present a promising avenue for accelerating DTI prediction compared to traditional experimental assays.
• The review highlights the diverse applications of DTI prediction, including drug discovery, adverse effect prediction, and drug repositioning.

Conclusions:

• Machine learning-based DTI prediction is a key strategy for streamlining the early stages of drug development.
• Computational approaches, particularly machine learning, are essential for overcoming the limitations of experimental DTI identification.
• Further research and application of these methods can significantly shorten the time to market for new therapeutics.