Regio-selectivity prediction with a machine-learned reaction representation and on-the-fly quantum mechanical descriptors.
Regio-selectivity prediction with a machine-learned reaction representation and on-the-fly quantum mechanical descriptors.
复制标题
用机器学习的反应表示和直播量子机械描述符的区域选择性预测。
DOI：
10.1039/d0sc04823b
复制
发表时间：
2020-12-22
期刊：
影响因子：
8.4
通讯作者：
Jensen KF
中科院分区：
文献类型：
--
作者：
Guan Y;Coley CW;Wu H;Ranasinghe D;Heid E;Struble TJ;Pattanaik L;Green WH;Jensen KF
准确而快速地评估基质是否能够经历所需的转变对于人类知识和计算机预测来说至关重要且具有挑战性。尽管机器学习在预测化学反应性（例如选择性）方面具有潜力，但流行的特征工程和学习方法要么耗时，要么需要大量数据。我们引入了一种新方法，将机器学习的反应表示与选定的量子力学描述符相结合，以预测一般取代反应中的区域选择性。我们基于 130k 有机分子的从头计算构建了反应性描述符数据库，并训练多任务约束模型来即时计算所需的描述符。所提出的平台增强了区域选择性预测的内推/外推性能，并能够从仅包含数百个示例的小型数据集中进行学习。此外，所提出的协议被证明普遍适用于各种化学空间。对于从商业数据库中挑选的三种常见类型的取代反应（芳香族 C-H 官能化、芳香族 C-X 取代和其他取代反应），融合模型在预测主要结果方面分别达到了 89.7%、96.7% 和 97.2% 的 top-1 准确率，每种反应均使用 5000 个训练反应。使用预测描述符，融合模型是端到端的，并且每个反应仅需要大约 70 毫秒来预测反应 SMILES 字符串的选择性。将特征学习和动态羽毛工程相结合，可以使用大型或小型数据集进行快速、准确的反应性预测。
Accurate and rapid evaluation of whether substrates can undergo the desired the transformation is crucial and challenging for both human knowledge and computer predictions. Despite the potential of machine learning in predicting chemical reactivity such as selectivity, popular feature engineering and learning methods are either time-consuming or data-hungry. We introduce a new method that combines machine-learned reaction representation with selected quantum mechanical descriptors to predict regio-selectivity in general substitution reactions. We construct a reactivity descriptor database based on ab initio calculations of 130k organic molecules, and train a multi-task constrained model to calculate demanded descriptors on-the-fly. The proposed platform enhances the inter/extra-polated performance for regio-selectivity predictions and enables learning from small datasets with just hundreds of examples. Furthermore, the proposed protocol is demonstrated to be generally applicable to a diverse range of chemical spaces. For three general types of substitution reactions (aromatic C–H functionalization, aromatic C–X substitution, and other substitution reactions) curated from a commercial database, the fusion model achieves 89.7%, 96.7%, and 97.2% top-1 accuracy in predicting the major outcome, respectively, each using 5000 training reactions. Using predicted descriptors, the fusion model is end-to-end, and requires approximately only 70 ms per reaction to predict the selectivity from reaction SMILES strings. Integrating feature learning and on-the-fly feather engineering enables fast and accurate reacitvity predictions using large or small dataset.
登录
查看更多内容
影响因子：
16.6
作者：
Jaworski, Wojciech;Szymkuc, Sara;Grzybowski, Bartosz A.
通讯作者：
Grzybowski, Bartosz A.
影响因子：
16.6
作者：
Beker, Wiktor;Gajewska, Ewa P.;Grzybowski, Bartosz A.
通讯作者：
Grzybowski, Bartosz A.
影响因子：
5.5
作者：
Heid, Esther;Fleck, Markus;MacKerell, Alexander D., Jr.
通讯作者：
MacKerell, Alexander D., Jr.
影响因子：
4.4
作者：
BECKE, AD
通讯作者：
BECKE, AD
影响因子：
7.3
作者：
Cherkasov, Artem;Muratov, Eugene N.;Fourches, Denis;Varnek, Alexandre;Baskin, Igor I.;Cronin, Mark;Dearden, John;Gramatica, Paola;Martin, Yvonne C.;Todeschini, Roberto;Consonni, Viviana;Kuz'min, Victor E.;Cramer, Richard;Benigni, Romualdo;Yang, Chihae;Rathman, James;Terfloth, Lothar;Gasteiger, Johann;Richard, Ann;Tropsha, Alexander
通讯作者：
Tropsha, Alexander