Drug-Target Interaction Prediction Techniques: A Review Article

کد مقاله : 1163-NAEC

نویسندگان

مریم طاهری *، محمدرضا کیوان پور دانشگاه الزهرا

چکیده مقاله

Predicting drug-target interactions (DTIs) plays a vital role in accelerating drug discovery and repositioning processes. Given the high cost and time required for experimental screening, modern methods have become indispensable. This paper traces the progression from traditional structure-based and ligand-based approaches to the advent of modern machine learning techniques. Moreover, the growing trend of integrating diverse data sources and multi-modal information is explored, along with challenges such as data scarcity, interpretability, and model generalization. This review presents a structured summary of the current landscape of DTI prediction, organizing various methods into five primary categories: computational, deep learning-based, graph-based, matrix-based, and multi-modal-based approaches. The goal is to provide a comprehensive framework for understanding these methodologies, highlighting key challenges, and suggesting future research directions. Future advancements will rely on the development of more generalizable and interpretable models, as well as effective use of the growing variety of biomedical data to uncover the complexities of DTI.

کلیدواژه ها

drug-target interaction, classical machine learning, deep learning, heterogeneous graph, matrix factorization

وضعیت: پذیرفته شده