{"title":"Conf-GEM:几何信息辅助直接构象生成模型","authors":"","doi":"10.1016/j.aichem.2024.100074","DOIUrl":null,"url":null,"abstract":"<div><p>Molecular conformations generation (MCG) aims to efficiently obtain reasonable and stable three-dimensional (3D) atomic coordinates of the atoms in the molecule from scratch, providing a structural foundation for molecular representation learning models and advanced downstream molecular design tasks such as molecular property prediction, molecular generation, and molecular docking. Existing MCG methods mostly rely on indirect distance-based strategies, which which can result in geometrically unrealistic conformations, or direct coordinate-based methods, which have larger search spaces and are prone to overfitting. Therefore, this study introduces Conf-GEM, a novel geometric information-assisted direct conformation generation model based on E-GeoGNN, a geometrically augmented 3D graph neural network with multiple scales. Pre-training and divide-and-conquer strategies, are integrated into the proposed model. Conf-GEM outperforms RDKit and nine deep-learning-based MCG models on the GEOM-QM9 and GEOM-Drugs datasets, achieving conformational coverage of 96.69% and 96.07%, respectively, without force field optimization. It also excels on the X-ray diffraction crystal structure dataset with up to 97.04% conformational coverage. In conclusion, Conf-GEM provides a novel solution for stabilizing 3D conformations generation. We provide an online prediction service (<span><span>https://confgem.cmdrg.com</span><svg><path></path></svg></span>) with a user-friendly interface for researchers.</p></div>","PeriodicalId":72302,"journal":{"name":"Artificial intelligence chemistry","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-07-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2949747724000320/pdfft?md5=48affbdd2252ef50c6eb12dedcdeacc7&pid=1-s2.0-S2949747724000320-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Conf-GEM: A geometric information-assisted direct conformation generation model\",\"authors\":\"\",\"doi\":\"10.1016/j.aichem.2024.100074\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Molecular conformations generation (MCG) aims to efficiently obtain reasonable and stable three-dimensional (3D) atomic coordinates of the atoms in the molecule from scratch, providing a structural foundation for molecular representation learning models and advanced downstream molecular design tasks such as molecular property prediction, molecular generation, and molecular docking. Existing MCG methods mostly rely on indirect distance-based strategies, which which can result in geometrically unrealistic conformations, or direct coordinate-based methods, which have larger search spaces and are prone to overfitting. Therefore, this study introduces Conf-GEM, a novel geometric information-assisted direct conformation generation model based on E-GeoGNN, a geometrically augmented 3D graph neural network with multiple scales. Pre-training and divide-and-conquer strategies, are integrated into the proposed model. Conf-GEM outperforms RDKit and nine deep-learning-based MCG models on the GEOM-QM9 and GEOM-Drugs datasets, achieving conformational coverage of 96.69% and 96.07%, respectively, without force field optimization. It also excels on the X-ray diffraction crystal structure dataset with up to 97.04% conformational coverage. In conclusion, Conf-GEM provides a novel solution for stabilizing 3D conformations generation. We provide an online prediction service (<span><span>https://confgem.cmdrg.com</span><svg><path></path></svg></span>) with a user-friendly interface for researchers.</p></div>\",\"PeriodicalId\":72302,\"journal\":{\"name\":\"Artificial intelligence chemistry\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-07-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S2949747724000320/pdfft?md5=48affbdd2252ef50c6eb12dedcdeacc7&pid=1-s2.0-S2949747724000320-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Artificial intelligence chemistry\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2949747724000320\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Artificial intelligence chemistry","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2949747724000320","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
分子构象生成(MCG)旨在从零开始有效地获得分子中原子的合理而稳定的三维(3D)原子坐标,为分子表征学习模型和高级下游分子设计任务(如分子性质预测、分子生成和分子对接)提供结构基础。现有的 MCG 方法大多依赖于基于间接距离的策略,这可能导致几何构象不切实际;或基于直接坐标的方法,其搜索空间较大,容易出现过拟合。因此,本研究介绍了一种基于 E-GeoGNN 的新型几何信息辅助直接构象生成模型 Conf-GEM,E-GeoGNN 是一种几何增强的多尺度三维图神经网络。预训练和分而治之策略被集成到了所提出的模型中。在 GEOM-QM9 和 GEOM-Drugs 数据集上,Conf-GEM 的表现优于 RDKit 和九种基于深度学习的 MCG 模型,在不进行力场优化的情况下,构象覆盖率分别达到 96.69% 和 96.07%。它在 X 射线衍射晶体结构数据集上也表现出色,构象覆盖率高达 97.04%。总之,Conf-GEM 为稳定三维构象的生成提供了一种新的解决方案。我们为研究人员提供了具有友好用户界面的在线预测服务(https://confgem.cmdrg.com)。
Conf-GEM: A geometric information-assisted direct conformation generation model
Molecular conformations generation (MCG) aims to efficiently obtain reasonable and stable three-dimensional (3D) atomic coordinates of the atoms in the molecule from scratch, providing a structural foundation for molecular representation learning models and advanced downstream molecular design tasks such as molecular property prediction, molecular generation, and molecular docking. Existing MCG methods mostly rely on indirect distance-based strategies, which which can result in geometrically unrealistic conformations, or direct coordinate-based methods, which have larger search spaces and are prone to overfitting. Therefore, this study introduces Conf-GEM, a novel geometric information-assisted direct conformation generation model based on E-GeoGNN, a geometrically augmented 3D graph neural network with multiple scales. Pre-training and divide-and-conquer strategies, are integrated into the proposed model. Conf-GEM outperforms RDKit and nine deep-learning-based MCG models on the GEOM-QM9 and GEOM-Drugs datasets, achieving conformational coverage of 96.69% and 96.07%, respectively, without force field optimization. It also excels on the X-ray diffraction crystal structure dataset with up to 97.04% conformational coverage. In conclusion, Conf-GEM provides a novel solution for stabilizing 3D conformations generation. We provide an online prediction service (https://confgem.cmdrg.com) with a user-friendly interface for researchers.