关键词: N6-甲基腺嘌呤修饰, 修饰检测方法, 修饰位点预测, 人工智能算法, 分子动力学模拟

Abstract: N6-methyladenosine (m6A) is the most prevalent modification in eukaryotic mRNA, playing a pivotal role in regu-lating various aspects of mRNA metabolism, including splicing, processing, degradation, and translation. This re-view provides a comprehensive overview of computational strategies employed in m6A research, with an emphasis on data-driven methodologies for the prediction of m6A sites and molecular dynamics simulations for deciphering m6A-associated biological mechanisms. The article first discusses the evolution of m6A detection technologies, out-lines the corresponding data processing methods, and summarizes publicly available datasets that serve as essential resources for constructing computational models. Subsequently, we highlight research advancements in machine learning and deep learning models for m6A site prediction. Finally, we demonstrate the contributions of molecular dynamics simulations in unravelling m6A-related molecular mechanisms, illustrating how computational methods facilitate the understanding of this complex epigenetic regulation. By systematically synthesizing relevant content, this review further discusses the latest research progress and application values of computational methods in m6A modification, offering new perspectives and insights for in-depth investigations.

Key words: N6-methyladenosine modification, modification detection method, modification sites prediction, artificial intelli-gence algorithm, molecular dynamic simulation