FragGen: towards 3D geometry reliable fragment-based molecular generation

被引:1
|
作者
Zhang, Odin [1 ]
Huang, Yufei [2 ]
Cheng, Shichen [1 ]
Yu, Mengyao [1 ]
Zhang, Xujun [1 ]
Lin, Haitao [2 ]
Zeng, Yundian [1 ]
Wang, Mingyang [1 ]
Wu, Zhenxing [1 ]
Zhao, Huifeng [1 ]
Zhang, Zaixi [3 ]
Hua, Chenqing [4 ]
Kang, Yu [1 ]
Cui, Sunliang [1 ]
Pan, Peichen [1 ]
Hsieh, Chang-Yu [1 ]
Hou, Tingjun [1 ]
机构
[1] Zhejiang Univ, Coll Pharmaceut Sci, Hangzhou 310058, Zhejiang, Peoples R China
[2] Zhejiang Univ, Hangzhou 310058, Zhejiang, Peoples R China
[3] Univ Sci & Technol China, Anhui Prov Key Lab Big Data Anal & Applicat, Hefei, Anhui, Peoples R China
[4] McGill Univ, Montreal Inst Learning Algorithms, Montreal, PQ, Canada
基金
中国国家自然科学基金;
关键词
DRUG; INHIBITORS; DISCOVERY; EFFICIENT; DOCKING;
D O I
10.1039/d4sc04620j
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
3D structure-based molecular generation is a successful application of generative AI in drug discovery. Most earlier models follow an atom-wise paradigm, generating molecules with good docking scores but poor molecular properties (like synthesizability and drugability). In contrast, fragment-wise generation offers a promising alternative by assembling chemically viable fragments. However, the co-design of plausible chemical and geometrical structures is still challenging, as evidenced by existing models. To address this, we introduce the Deep Geometry Handling protocol, which decomposes the entire geometry into multiple sets of geometric variables, looking beyond model architecture design. Drawing from a newly defined six-category taxonomy, we propose FragGen, a novel hybrid strategy as the first geometry-reliable, fragment-wise molecular generation method. FragGen significantly enhances both the geometric quality and synthesizability of the generated molecules, overcoming major limitations of previous models. Moreover, FragGen has been successfully applied in real-world scenarios, notably in designing type II kinase inhibitors at the similar to nM level, establishing it as the first validated 3D fragment-based drug design algorithm. We believe that this concept-algorithm-application cycle will not only inspire researchers working on other geometry-centric tasks to move beyond architecture designs but also provide a solid example of how generative AI can be customized for drug design.
引用
收藏
页码:19452 / 19465
页数:14
相关论文
共 50 条
  • [11] Towards DRM for 3D geometry data
    Gschwandtner, Michael
    Uhl, Andreas
    SECURITY, FORENSICS, STEGANOGRAPHY, AND WATERMARKING OF MULTIMEDIA CONTENTS X, 2008, 6819
  • [12] Explainable Fragment-Based Molecular Property Attribution
    Jia, Lingxiang
    Feng, Zunlei
    Zhang, Haotian
    Song, Jie
    Zhong, Zipeng
    Yao, Shaolun
    Song, Mingli
    ADVANCED INTELLIGENT SYSTEMS, 2022, 4 (10)
  • [13] Fragon - rapid fragment-based molecular replacement
    Jenkins, Huw T.
    Antson, Alfred A.
    ACTA CRYSTALLOGRAPHICA A-FOUNDATION AND ADVANCES, 2016, 72 : S22 - S23
  • [14] Fragment-based methods for large molecular systems
    Saha, Arjun
    Raghavachari, Krishnan
    ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, 2013, 246
  • [15] Fragment-based Pretraining and Finetuning on Molecular Graphs
    Kha-Dinh Luong
    Singh, Ambuj
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [16] e-Drug3D: 3D structure collections dedicated to drug repurposing and fragment-based drug design
    Pihan, Emilie
    Colliandre, Lionel
    Guichou, Jean-Francois
    Douguet, Dominique
    BIOINFORMATICS, 2012, 28 (11) : 1540 - 1541
  • [17] 3D structure generation based on distance geometry algorithm.
    Nakamura, K
    Tomika, N
    Itai, A
    ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, 2002, 223 : U493 - U493
  • [18] A Deep Generative Model for Fragment-Based Molecule Generation
    Podda, Marco
    Bacciu, Davide
    Micheli, Alessio
    INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 108, 2020, 108 : 2240 - 2249
  • [19] An improved model for fragment-based lead generation at AstraZeneca
    Fuller, Nathan
    Spadola, Loredana
    Cowen, Scott
    Patel, Joe
    Schonherr, Heike
    Cao, Qing
    McKenzie, Andrew
    Edfeldt, Fredrik
    Rabow, Al
    Goodnow, Robert
    DRUG DISCOVERY TODAY, 2016, 21 (08) : 1272 - 1283
  • [20] Expanding medicinal chemistry into 3D space: metallofragments as 3D scaffolds for fragment-based drug discovery (vol 11, pg 1216, 2020)
    Morrison, Christine N.
    Prosser, Kathleen E.
    Stokes, Ryjul W.
    Cordes, Anna
    Metzler-Nolte, Nils
    Cohen, Seth M.
    CHEMICAL SCIENCE, 2022, 13 (32) : 9450 - 9452