1
|
Chen S, Xie J, Ye R, Xu DD, Yang Y. Structure-aware dual-target drug design through collaborative learning of pharmacophore combination and molecular simulation. Chem Sci 2024; 15:10366-10380. [PMID: 38994407 PMCID: PMC11234869 DOI: 10.1039/d4sc00094c] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/05/2024] [Accepted: 06/09/2024] [Indexed: 07/13/2024] Open
Abstract
Dual-target drug design has gained significant attention in the treatment of complex diseases, such as cancers and autoimmune disorders. A widely employed design strategy is combining pharmacophores to leverage the knowledge of structure-activity relationships of both targets. Unfortunately, pharmacophore combination often struggles with long and expensive trial and error, because the protein pockets of the two targets impose complex structural constraints. In this study, we propose AIxFuse, a structure-aware dual-target drug design method that learns pharmacophore fusion patterns to satisfy the dual-target structural constraints simulated by molecular docking. AIxFuse employs two self-play reinforcement learning (RL) agents to learn pharmacophore selection and fusion by comprehensive feedback including dual-target molecular docking scores. Collaboratively, the molecular docking scores are learned by active learning (AL). Through collaborative RL and AL, AIxFuse learns to generate molecules with multiple desired properties. AIxFuse is shown to outperform state-of-the-art methods in generating dual-target drugs against glycogen synthase kinase-3 beta (GSK3β) and c-Jun N-terminal kinase 3 (JNK3). When applied to another task against retinoic acid receptor-related orphan receptor γ-t (RORγt) and dihydroorotate dehydrogenase (DHODH), AIxFuse exhibits consistent performance while compared methods suffer from performance drops, leading to a 5 times higher performance in success rate. Docking studies demonstrate that AIxFuse can generate molecules concurrently satisfying the binding mode required by both targets. Further free energy perturbation calculation indicates that the generated candidates have promising binding free energies against both targets.
Collapse
Affiliation(s)
- Sheng Chen
- School of Computer Science and Engineering, Sun Yat-sen University Guangzhou 510006 China
- AixplorerBio Inc. Jiaxing 314031 China
| | - Junjie Xie
- School of Computer Science and Engineering, Sun Yat-sen University Guangzhou 510006 China
- AixplorerBio Inc. Jiaxing 314031 China
| | | | | | - Yuedong Yang
- School of Computer Science and Engineering, Sun Yat-sen University Guangzhou 510006 China
| |
Collapse
|
2
|
Ahmad S, Raza K. An extensive review on lung cancer therapeutics using machine learning techniques: state-of-the-art and perspectives. J Drug Target 2024; 32:635-646. [PMID: 38662768 DOI: 10.1080/1061186x.2024.2347358] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/10/2024] [Accepted: 04/18/2024] [Indexed: 05/07/2024]
Abstract
There are over 100 types of human cancer, accounting for millions of deaths every year. Lung cancer alone claims over 1.8 million lives per year and is expected to surpass 3.2 million by 2050, which underscores the urgent need for rapid drug development and repurposing initiatives. The application of AI emerges as a pivotal solution to developing anti-cancer therapeutics. This state-of-the-art review aims to explore the various applications of AI in lung cancer therapeutics. Predictive models can analyse large datasets, including clinical data, genetic information, and treatment outcomes, for novel drug design and to generate personalised treatment recommendations, potentially optimising therapeutic strategies, enhancing treatment efficacy, and minimising adverse effects. A thorough literature review study was conducted based on articles indexed in PubMed and Scopus. We compiled the use of various machine learning approaches, including CNN, RNN, GAN, VAEs, and other AI techniques, enhancing efficiency with accuracy exceeding 95%, which is validated through a computer-aided drug design process. AI can revolutionise lung cancer therapeutics, streamlining processes and saving biological scientists' time and effort-however, further research is needed to overcome challenges and fully unlock AI's potential in Lung Cancer Therapeutics.
Collapse
Affiliation(s)
- Shaban Ahmad
- Department of Computer Science, Jamia Millia Islamia, New Delhi, India
| | - Khalid Raza
- Department of Computer Science, Jamia Millia Islamia, New Delhi, India
| |
Collapse
|
3
|
Xie J, Chen S, Lei J, Yang Y. DiffDec: Structure-Aware Scaffold Decoration with an End-to-End Diffusion Model. J Chem Inf Model 2024; 64:2554-2564. [PMID: 38267393 DOI: 10.1021/acs.jcim.3c01466] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/26/2024]
Abstract
In molecular optimization, one popular way is R-group decoration on molecular scaffolds, and many efforts have been made to generate R-groups based on deep generative models. However, these methods mostly use information on known binding ligands, without fully utilizing target structure information. In this study, we proposed a new method, DiffDec, to involve 3D pocket constraints by a modified diffusion technique for optimizing molecules through molecular scaffold decoration. For end-to-end generation of R-groups with different sizes, we designed a novel fake atom mechanism. DiffDec was shown to be able to generate structure-aware R-groups with realistic geometric substructures by the analysis of bond angles and dihedral angles and simultaneously generate multiple R-groups for one scaffold on different growth anchors. The growth anchors could be provided by users or automatically determined by our model. DiffDec achieved R-group recovery rates of 69.67% and 45.34% in the single and multiple R-group decoration tasks, respectively, and these values were significantly higher than competing methods (37.33% and 26.85%). According to the molecular docking study, our decorated molecules obtained a better average binding affinity than baseline methods. The docking pose analysis revealed that DiffDec could decorate scaffolds with R-groups that exhibited improved binding affinities and more favorable interactions with the pocket. These results demonstrated the potential and applicability of DiffDec in real-world scaffold decoration for molecular optimization.
Collapse
Affiliation(s)
- Junjie Xie
- School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510006, China
- AixplorerBio Inc., Jiaxing 314031, China
| | - Sheng Chen
- School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510006, China
- AixplorerBio Inc., Jiaxing 314031, China
| | - Jinping Lei
- School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou 510006, China
| | - Yuedong Yang
- School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510006, China
| |
Collapse
|
4
|
Zhang H, Huang J, Xie J, Huang W, Yang Y, Xu M, Lei J, Chen H. GRELinker: A Graph-Based Generative Model for Molecular Linker Design with Reinforcement and Curriculum Learning. J Chem Inf Model 2024; 64:666-676. [PMID: 38241022 DOI: 10.1021/acs.jcim.3c01700] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/13/2024]
Abstract
Fragment-based drug discovery (FBDD) is widely used in drug design. One useful strategy in FBDD is designing linkers for linking fragments to optimize their molecular properties. In the current study, we present a novel generative fragment linking model, GRELinker, which utilizes a gated-graph neural network combined with reinforcement and curriculum learning to generate molecules with desirable attributes. The model has been shown to be efficient in multiple tasks, including controlling log P, optimizing synthesizability or predicted bioactivity of compounds, and generating molecules with high 3D similarity but low 2D similarity to the lead compound. Specifically, our model outperforms the previously reported reinforcement learning (RL) built-in method DRlinker on these benchmark tasks. Moreover, GRELinker has been successfully used in an actual FBDD case to generate optimized molecules with enhanced affinities by employing the docking score as the scoring function in RL. Besides, the implementation of curriculum learning in our framework enables the generation of structurally complex linkers more efficiently. These results demonstrate the benefits and feasibility of GRELinker in linker design for molecular optimization and drug discovery.
Collapse
Affiliation(s)
- Hao Zhang
- School of Pharmaceutical Science, Sun Yat-sen University, Guangzhou 510006, China
| | - Jinchao Huang
- School of Pharmaceutical Science, Sun Yat-sen University, Guangzhou 510006, China
| | - Junjie Xie
- School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510006, China
| | - Weifeng Huang
- School of Pharmaceutical Science, Sun Yat-sen University, Guangzhou 510006, China
| | - Yuedong Yang
- School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510006, China
| | - Mingyuan Xu
- Guangzhou National Laboratory, Guangzhou International Bio Island, No. 9 Xin Dao Huan Bei Road, Guangzhou 510005, China
| | - Jinping Lei
- School of Pharmaceutical Science, Sun Yat-sen University, Guangzhou 510006, China
| | - Hongming Chen
- Guangzhou National Laboratory, Guangzhou International Bio Island, No. 9 Xin Dao Huan Bei Road, Guangzhou 510005, China
| |
Collapse
|
5
|
Zha J, He J, Wu C, Zhang M, Liu X, Zhang J. Designing drugs and chemical probes with the dualsteric approach. Chem Soc Rev 2023; 52:8651-8677. [PMID: 37990599 DOI: 10.1039/d3cs00650f] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/23/2023]
Abstract
Traditionally, drugs are monovalent, targeting only one site on the protein surface. This includes orthosteric and allosteric drugs, which bind the protein at orthosteric and allosteric sites, respectively. Orthosteric drugs are good in potency, whereas allosteric drugs have better selectivity and are solutions to classically undruggable targets. However, it would be difficult to simultaneously reach high potency and selectivity when targeting only one site. Also, both kinds of monovalent drugs suffer from mutation-caused drug resistance. To overcome these obstacles, dualsteric modulators have been proposed in the past twenty years. Compared to orthosteric or allosteric drugs, dualsteric modulators are bivalent (or bitopic) with two pharmacophores. Each of the two pharmacophores bind the protein at the orthosteric and an allosteric site, which could bring the modulator with special properties beyond monovalent drugs. In this study, we comprehensively review the current development of dualsteric modulators. Our main effort reason and illustrate the aims to apply the dualsteric approach, including a "double win" of potency and selectivity, overcoming mutation-caused drug resistance, developments of function-biased modulators, and design of partial agonists. Moreover, the strengths of the dualsteric technique also led to its application outside pharmacy, including the design of highly sensitive fluorescent tracers and usage as molecular rulers. Besides, we also introduced drug targets, designing strategies, and validation methods of dualsteric modulators. Finally, we detail the conclusions and perspectives.
Collapse
Affiliation(s)
- Jinyin Zha
- College of Pharmacy, Ningxia Medical University, Yinchuan, Ningxia Hui Autonomous Region, China.
- State Key Laboratory of Medical Genomics, National Research Center for Translational Medicine at Shanghai, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
| | - Jixiao He
- State Key Laboratory of Medical Genomics, National Research Center for Translational Medicine at Shanghai, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
| | - Chengwei Wu
- State Key Laboratory of Medical Genomics, National Research Center for Translational Medicine at Shanghai, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
| | - Mingyang Zhang
- State Key Laboratory of Medical Genomics, National Research Center for Translational Medicine at Shanghai, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
| | - Xinyi Liu
- College of Pharmacy, Ningxia Medical University, Yinchuan, Ningxia Hui Autonomous Region, China.
- State Key Laboratory of Medical Genomics, National Research Center for Translational Medicine at Shanghai, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
| | - Jian Zhang
- College of Pharmacy, Ningxia Medical University, Yinchuan, Ningxia Hui Autonomous Region, China.
- State Key Laboratory of Medical Genomics, National Research Center for Translational Medicine at Shanghai, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
- Medicinal Chemistry and Bioinformatics Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China
| |
Collapse
|
6
|
Xu M, Chen H. Tree-Invent: A Novel Multipurpose Molecular Generative Model Constrained with a Topological Tree. J Chem Inf Model 2023; 63:7067-7082. [PMID: 37962855 DOI: 10.1021/acs.jcim.3c01626] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/15/2023]
Abstract
De novo molecular design plays an important role in drug discovery. Here, a novel generative model, Tree-Invent, was proposed to integrate topological constraints in the generation of a molecular graph. In this model, a molecular graph is represented as a topological tree in which a ring system, a nonring atom, and a chemical bond are regarded as the ring node, single node, and edge, respectively. The molecule generation is driven by three independent submodels for carrying out operations of node addition, ring generation, and node connection. One unique feature of the generative model is that the topological tree structure can be specified as a constraint for structure generation, which provides more precise control of structure generation. Combined with reinforcement learning, the Tree-Invent model could efficiently explore targeted chemical space. Moreover, the Tree-Invent model is flexible enough to be used in versatile molecule design settings such as scaffold decoration, scaffold hopping, and linker generation.
Collapse
Affiliation(s)
- Mingyuan Xu
- Guangzhou National Laboratory, No. 9 XingDaoHuanBei Road, Guangzhou International Bio Island, Guangzhou, Guangdong 510005, China
| | - Hongming Chen
- Guangzhou National Laboratory, No. 9 XingDaoHuanBei Road, Guangzhou International Bio Island, Guangzhou, Guangdong 510005, China
| |
Collapse
|
7
|
Danishuddin, Jamal MS, Song KS, Lee KW, Kim JJ, Park YM. Revolutionizing Drug Targeting Strategies: Integrating Artificial Intelligence and Structure-Based Methods in PROTAC Development. Pharmaceuticals (Basel) 2023; 16:1649. [PMID: 38139776 PMCID: PMC10747325 DOI: 10.3390/ph16121649] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/24/2023] [Revised: 11/20/2023] [Accepted: 11/22/2023] [Indexed: 12/24/2023] Open
Abstract
PROteolysis TArgeting Chimera (PROTAC) is an emerging technology in chemical biology and drug discovery. This technique facilitates the complete removal of the target proteins that are "undruggable" or challenging to target through chemical molecules via the Ubiquitin-Proteasome System (UPS). PROTACs have been widely explored and outperformed not only in cancer but also in other diseases. During the past few decades, several academic institutes and pharma companies have poured more efforts into PROTAC-related technologies, setting the stage for several major degrader trial readouts in clinical phases. Despite their promising results, the formation of robust ternary orientation, off-target activity, poor permeability, and binding affinity are some of the limitations that hinder their development. Recent advancements in computational technologies have facilitated progress in the development of PROTACs. Researchers have been able to utilize these technologies to explore a wider range of E3 ligases and optimize linkers, thereby gaining a better understanding of the effectiveness and safety of PROTACs in clinical settings. In this review, we briefly explore the computational strategies reported to date for the formation of PROTAC components and discuss the key challenges and opportunities for further research in this area.
Collapse
Affiliation(s)
- Danishuddin
- Department of Biotechnology, Yeungnam University, Gyeongsan 38541, Republic of Korea;
| | | | - Kyoung-Seob Song
- Department of Medical Science, Kosin University College of Medicine, 194 Wachi-ro, Yeongdo-gu, Busan 49104, Republic of Korea;
| | - Keun-Woo Lee
- Division of Life Science, Department of Bio & Medical Big-Data (BK4 Program), Research Institute of Natural Science (RINS), Gyeongsang National University (GNU), 501 Jinju-daero, Jinju 52828, Republic of Korea
- Angel i-Drug Design (AiDD), 33-3 Jinyangho-ro 44, Jinju 52650, Republic of Korea
| | - Jong-Joo Kim
- Department of Biotechnology, Yeungnam University, Gyeongsan 38541, Republic of Korea;
| | - Yeong-Min Park
- Department of Integrative Biological Sciences and Industry, Sejong University, 209, Neugdong-ro, Gwangjin-gu, Seoul 05006, Republic of Korea
| |
Collapse
|
8
|
Patterson-Gardner C, Pavelich GM, Cannon AT, Menke AJ, Simanek EE. Adaptation of Empirical Methods to Predict the LogD of Triazine Macrocycles. ACS Med Chem Lett 2023; 14:1378-1382. [PMID: 37849549 PMCID: PMC10577694 DOI: 10.1021/acsmedchemlett.3c00290] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/03/2023] [Accepted: 09/01/2023] [Indexed: 10/19/2023] Open
Abstract
Octanol/water partition coefficients guide drug design, but algorithms do not always accurately predict these values. For cationic triazine macrocycles that adopt a conserved folded shape in solution, common algorithms fall short. Here, the logD values for 12 macrocycles differing in amino acid choice were predicted and then measured experimentally. On average, AlogP, XlogP, and ChemAxon predictions deviate by 0.9, 2.8, and 3.9 log units, with XlogP overestimating lipophilicity and AlogP and ChemAxon underestimating lipophilicity. Importantly, however, a linear relationship (R2 > 0.98) exists between the values predicted by AlogP and the experimentally determined logD values, thus enabling more accurate predictions.
Collapse
Affiliation(s)
- Casey
J. Patterson-Gardner
- Department of Chemistry & Biochemistry, Texas Christian University, Fort Worth, Texas 76129, United States
| | - Gretchen M. Pavelich
- Department of Chemistry & Biochemistry, Texas Christian University, Fort Worth, Texas 76129, United States
| | - April T. Cannon
- Department of Chemistry & Biochemistry, Texas Christian University, Fort Worth, Texas 76129, United States
| | - Alexander J. Menke
- Department of Chemistry & Biochemistry, Texas Christian University, Fort Worth, Texas 76129, United States
| | - Eric E. Simanek
- Department of Chemistry & Biochemistry, Texas Christian University, Fort Worth, Texas 76129, United States
| |
Collapse
|
9
|
Hagg A, Kirschner KN. Open-Source Machine Learning in Computational Chemistry. J Chem Inf Model 2023; 63:4505-4532. [PMID: 37466636 PMCID: PMC10430767 DOI: 10.1021/acs.jcim.3c00643] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/04/2023] [Indexed: 07/20/2023]
Abstract
The field of computational chemistry has seen a significant increase in the integration of machine learning concepts and algorithms. In this Perspective, we surveyed 179 open-source software projects, with corresponding peer-reviewed papers published within the last 5 years, to better understand the topics within the field being investigated by machine learning approaches. For each project, we provide a short description, the link to the code, the accompanying license type, and whether the training data and resulting models are made publicly available. Based on those deposited in GitHub repositories, the most popular employed Python libraries are identified. We hope that this survey will serve as a resource to learn about machine learning or specific architectures thereof by identifying accessible codes with accompanying papers on a topic basis. To this end, we also include computational chemistry open-source software for generating training data and fundamental Python libraries for machine learning. Based on our observations and considering the three pillars of collaborative machine learning work, open data, open source (code), and open models, we provide some suggestions to the community.
Collapse
Affiliation(s)
- Alexander Hagg
- Institute
of Technology, Resource and Energy-Efficient Engineering (TREE), University of Applied Sciences Bonn-Rhein-Sieg, 53757 Sankt Augustin, Germany
- Department
of Electrical Engineering, Mechanical Engineering and Technical Journalism, University of Applied Sciences Bonn-Rhein-Sieg, 53757 Sankt Augustin, Germany
| | - Karl N. Kirschner
- Institute
of Technology, Resource and Energy-Efficient Engineering (TREE), University of Applied Sciences Bonn-Rhein-Sieg, 53757 Sankt Augustin, Germany
- Department
of Computer Science, University of Applied
Sciences Bonn-Rhein-Sieg, 53757 Sankt Augustin, Germany
| |
Collapse
|
10
|
Grenier D, Audebert S, Preto J, Guichou JF, Krimm I. Linkers in fragment-based drug design: an overview of the literature. Expert Opin Drug Discov 2023; 18:987-1009. [PMID: 37466331 DOI: 10.1080/17460441.2023.2234285] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/15/2023] [Accepted: 07/05/2023] [Indexed: 07/20/2023]
Abstract
INTRODUCTION In fragment-based drug design, fragment linking is a popular strategy where two fragments binding to different sub-pockets of a target are linked together. This attractive method remains challenging especially due to the design of ideal linkers. AREAS COVERED The authors review the types of linkers and chemical reactions commonly used to the synthesis of linkers, including those utilized in protein-templated fragment self-assembly, where fragments are directly linked in the presence of the protein. Finally, they detail computational workflows and software including generative models that have been developed for fragment linking. EXPERT OPINION The authors believe that fragment linking offers key advantages for compound design, particularly for the design of bivalent inhibitors linking two distinct pockets of the same or different subunits. On the other hand, more studies are needed to increase the potential of protein-templated approaches in FBDD. Important computational tools such as structure-based de novo software are emerging to select suitable linkers. Fragment linking will undoubtedly benefit from developments in computational approaches and machine learning models.
Collapse
Affiliation(s)
- Dylan Grenier
- Team Small Molecules for Biological Targets, Centre de Recherche En Cancérologie (CRCL) - INSERM 1052 - CNRS 5286 - Centre Léon Bérard - Université Claude Bernard Lyon 1, Institut Convergence Plascan, Lyon, France
| | - Solène Audebert
- Centre de Biologie Structurale, CNRS, INSERM, Univ. Montpellier, Montpellier, France
| | - Jordane Preto
- Team Small Molecules for Biological Targets, Centre de Recherche En Cancérologie (CRCL) - INSERM 1052 - CNRS 5286 - Centre Léon Bérard - Université Claude Bernard Lyon 1, Institut Convergence Plascan, Lyon, France
| | - Jean-François Guichou
- Centre de Biologie Structurale, CNRS, INSERM, Univ. Montpellier, Montpellier, France
| | - Isabelle Krimm
- Team Small Molecules for Biological Targets, Centre de Recherche En Cancérologie (CRCL) - INSERM 1052 - CNRS 5286 - Centre Léon Bérard - Université Claude Bernard Lyon 1, Institut Convergence Plascan, Lyon, France
| |
Collapse
|
11
|
Yamada M, Sugiyama M. Molecular Graph Generation by Decomposition and Reassembling. ACS OMEGA 2023; 8:19575-19586. [PMID: 37305268 PMCID: PMC10249382 DOI: 10.1021/acsomega.3c01078] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 02/17/2023] [Accepted: 04/10/2023] [Indexed: 06/13/2023]
Abstract
Designing molecular structures with desired chemical properties is an essential task in drug discovery and materials design. However, finding molecules with the optimized desired properties is still a challenging task due to combinatorial explosion of the candidate space of molecules. Here we propose a novel decomposition-and-reassembling-based approach, which does not include any optimization in hidden space, and our generation process is highly interpretable. Our method is a two-step procedure: In the first decomposition step, we apply frequent subgraph mining to a molecular database to collect a smaller size of subgraphs as building blocks of molecules. In the second reassembling step, we search desirable building blocks guided via reinforcement learning and combine them to generate new molecules. Our experiments show that our method not only can find better molecules in terms of two standard criteria, the penalized log P and druglikeness, but also can generate drug molecules showing the valid intermediate molecules.
Collapse
Affiliation(s)
- Masatsugu Yamada
- School
of Multidisciplinary Sciences, Department of Informatics, The Graduate University for Advanced Studies, SOKENDAI, Kanagawa 240-0115, Japan
- National
Institute of Informatics, Chiyoda-ku, Tokyo 101-8430, Japan
- Innovative
Technology Laboratories, AGC Inc., 230-0045 Kanagawa, Japan
| | - Mahito Sugiyama
- School
of Multidisciplinary Sciences, Department of Informatics, The Graduate University for Advanced Studies, SOKENDAI, Kanagawa 240-0115, Japan
- National
Institute of Informatics, Chiyoda-ku, Tokyo 101-8430, Japan
| |
Collapse
|
12
|
Yu Y, Huang J, He H, Han J, Ye G, Xu T, Sun X, Chen X, Ren X, Li C, Li H, Huang W, Liu Y, Wang X, Gao Y, Cheng N, Guo N, Chen X, Feng J, Hua Y, Liu C, Zhu G, Xie Z, Yao L, Zhong W, Chen X, Liu W, Li H. Accelerated Discovery of Macrocyclic CDK2 Inhibitor QR-6401 by Generative Models and Structure-Based Drug Design. ACS Med Chem Lett 2023; 14:297-304. [PMID: 36923916 PMCID: PMC10009793 DOI: 10.1021/acsmedchemlett.2c00515] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/08/2022] [Accepted: 01/19/2023] [Indexed: 02/11/2023] Open
Abstract
Selective CDK2 inhibitors have the potential to provide effective therapeutics for CDK2-dependent cancers and for combating drug resistance due to high cyclin E1 (CCNE1) expression intrinsically or CCNE1 amplification induced by treatment of CDK4/6 inhibitors. Generative models that take advantage of deep learning are being increasingly integrated into early drug discovery for hit identification and lead optimization. Here we report the discovery of a highly potent and selective macrocyclic CDK2 inhibitor QR-6401 (23) accelerated by the application of generative models and structure-based drug design (SBDD). QR-6401 (23) demonstrated robust antitumor efficacy in an OVCAR3 ovarian cancer xenograft model via oral administration.
Collapse
Affiliation(s)
- Yang Yu
- Tencent
AI Lab, Tencent, Shenzhen 518057, China
| | | | - Hu He
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Jing Han
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Geyan Ye
- Tencent
AI Lab, Tencent, Shenzhen 518057, China
| | - Tingyang Xu
- Tencent
AI Lab, Tencent, Shenzhen 518057, China
| | | | - Xiumei Chen
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Xiaoming Ren
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Chunlai Li
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Huijuan Li
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Wei Huang
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Yangyang Liu
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Xinjuan Wang
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Yongzhi Gao
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Nianhe Cheng
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Na Guo
- BioDuro-Sundia, Shanghai, 200131, China
| | - Xibo Chen
- BioDuro-Sundia, Shanghai, 200131, China
| | | | - Yuxia Hua
- BioDuro-Sundia, Beijing, 102200, China
| | - Chong Liu
- BioDuro-Sundia, Beijing, 102200, China
| | - Guoyun Zhu
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Zhi Xie
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Lili Yao
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Wenge Zhong
- Regor
Therapeutics Group, Shanghai, 201210, China
| | - Xinde Chen
- Tencent
AI Lab, Tencent, Shenzhen 518057, China
| | - Wei Liu
- Tencent
AI Lab, Tencent, Shenzhen 518057, China
| | - Hailong Li
- Regor
Therapeutics Group, Shanghai, 201210, China
| |
Collapse
|