Artificial Intelligence-Based Prediction of Drug-Target Interactions in Mental Illnesses

Abstract

Mental illnesses are a serious problem worldwide, and many aspects of their disease mechanisms and drug development remain unclear. Traditional drug development processes face several major challenges, including slow speed, high costs, and difficulty in achieving successful cures. However, drug-target interaction based on artificial intelligence is gradually becoming a research hotspot due to its advantages such as accessibility and low cost. This article reviews drug-target interaction research in the development of drugs for mental illnesses. It systematically summarizes and analyzes relevant research progress from three aspects: traditional biomedical methods, machine learning methods, and commonly used datasets, comparing the basic principles, application characteristics, advantages, and limitations of different methods. Based on this, it summarizes current research problems such as insufficient data quality, limited model generalization ability, insufficient interpretability, and insufficient depth in research targeting specific disease stages. Finally, it looks forward to the future development direction of artificial intelligence technology in intelligent drug discovery for mental illnesses, aiming to provide theoretical reference and practical guidance for subsequent related research.