General Protocol for the Accurate Prediction of Molecular 13C/1H NMR Chemical Shifts via Machine Learning Augmented DFT
General Protocol for the Accurate Prediction of Molecular 13C/1H NMR Chemical Shifts via Machine Learning Augmented DFT
复制标题
DOI：
10.1021/acs.jcim.0c00388
复制
发表时间：
2020-08-24
影响因子：
5.6
通讯作者：
Glezakou, Vassiliki-Alexandra
中科院分区：
文献类型：
--
作者：
Gao, Peng;Zhang, Jun;Glezakou, Vassiliki-Alexandra
在可承受的计算成本下准确预测核磁共振化学位移对于实验研究中不同类型的结构分配是非常重要的。密度泛函理论（DFT）和含量规原子轨道（GIAO）是核磁共振计算中最常用的两种计算方法，但它们往往不能解决结构赋值中的歧义。在这里，我们提出了一种使用机器学习（ML）技术（DFT + ML）的新方法，该方法显着提高了C-13/H-1核磁共振化学位移预测对各种有机分子的准确性。可推广的DFT + ML模型的输入包含两个关键部分：一个是提供对化学环境的见解的向量，可以在不知道分子确切几何形状的情况下进行评估；另一个是dft计算的各向同性屏蔽常数。DFT + ML模型使用包含476个C-13和270个H-1实验化学位移的数据集进行训练。对于本文使用的DFT方法，预测和实验C-13/H-1化学位移误差的均方根偏差（rmsd）可小至2.10/0.18 ppm，远低于简单DFT方法（5.54/0.25 ppm）或DFT +线性回归（LR）方法（4.77/0.23 ppm）。与之前提出的两种核磁共振预测ML模型相比，它的最大绝对误差也更小。在两类有机分子（TIC10和风信子）上测试了DFT + ML模型的鲁棒性，其中正确的异构体被明确地分配给实验分子。总的来说，DFT + ML模型显示了在各种系统（包括立体异构体）中进行结构分配的希望，这些系统通常难以通过实验确定。
An accurate prediction of NMR chemical shifts at affordable computational cost is very important for different types of structural assignments in experimental studies. Density functional theory (DFT) and gauge-including atomic orbital (GIAO) are two of the most popular computational methods for NMR calculation, yet they often fail to resolve ambiguities in structural assignments. Here, we present a new method that uses machine learning (ML) techniques (DFT + ML) that significantly increases the accuracy of C-13/H-1 NMR chemical shift prediction for a variety of organic molecules. The input of the generalizable DFT + ML model contains two critical parts: one is a vector providing insights into chemical environments, which can be evaluated without knowing the exact geometry of the molecule; the other one is the DFT-calculated isotropic shielding constant. The DFT + ML model was trained with a data set containing 476 C-13 and 270 H-1 experimental chemical shifts. For the DFT methods used here, the root mean square deviations (RMSDs) for the errors between predicted and experimental C-13/H-1 chemical shifts can be as small as 2.10/0.18 ppm, which is much lower than those from simple DFT (5.54/0.25 ppm), or DFT + linear regression (LR) (4.77/0.23 ppm) approaches. It also has a smaller maximum absolute error than two previously proposed NMR-predicting ML models. The robustness of the DFT + ML model is tested on two classes of organic molecules (TIC10 and hyacinthacines), where the correct isomers were unambiguously assigned to the experimental ones. Overall, the DFT + ML model shows promise for structural assignments in a variety of systems, including stereoisomers, that are often challenging to determine experimentally.