International Journal of Medicine and Data
International Journal of Medicine and Data. 2025; 9: (1) ; 10.12208/j.ijmd.20250017 .
总浏览量: 134
华北理工大学 河北唐山
*通讯作者： 徐璐洋,单位：华北理工大学 河北唐山；
近年来深度学习方法在小分子药物与蛋白质靶标的亲和力（DTA）预测领域取得了显著进展，生成对抗网络（GAN）已被应用于该领域，但当前主流方法多局限于浅层GAN架构。本文综述了基于深度学习的DTA预测方法，探讨了其在提高预测精度和模型可解释性方面的潜力。文章首先回顾了传统方法和深度学习方法的优缺点，随后重点介绍了多头注意力机制和GAN的结合应用，最后总结了当前研究的挑战和未来发展方向。
Recent advances in deep learning have significantly improved small molecule-drug target affinity (DTA) prediction, with generative adversarial networks (GANs) emerging as a promising approach. However, current GAN-based methods primarily use shallow architectures. This paper reviews deep learning-based DTA prediction methods, examining their potential to enhance accuracy and interpretability. We compare traditional and deep learning approaches, highlight the integration of multi-head attention with GANs, and discuss current challenges and future directions.
[1] Zhang D, Zhang T, Zhu J, et al. Discovery of novel targets and me chanisms of MEK inhibitor Selumetinib for LGG treatment based on molecular docking and molecular dynamics simulation[J]. Journal of Molecular Modeling, 2022, 28(6): 138.
[2] Pahikkala T, Airola A, Pietilä S, et al. Toward more realistic drug–target interaction predictions[J]. Briefings in bioinformatics, 2015, 16(2): 325-337.
[3] Zhu Z, Yao Z, Zheng X, Qi G, Li Y, Mazur N, et al. Drug–target affinity prediction method based on multi-scale information interaction and graph optimization. Comput[J]. Biol, 2023, Med. 167.
[4] Thin Nguyen, Hang Le, Thomas P Quinn, Tri Nguyen, Thuc Duy Le, Svetha Venkatesh, GraphDTA: predicting drug–target binding affinity with graph neural networks, Bioinformatics, Volume 37, Issue 8, March 2021, Pages 1140–1147.
[5] 彭杨,朱小飞,胡冬冬.基于注意力感知和模态融合的药物-靶点相互作用预测方法[J/OL].太原理工大学学报:1-10[2024-04-23].