从使用神经网络的核磁共振光谱自动确定分子子结构
Shiyun Liu1, Jacqueline M Cole1,2
1Cavendish Laboratory, Department of Physics, University of Cambridge, J. J. Thomson Avenue, Cambridge CB3 0HE. U.K.
Journal of chemical information and modeling
|August 13, 2025
概括
机器学习模型现在可以自动解释核磁共振 (NMR) 光谱,加速分子结构的确定. 卷积神经网络 (CNN) 为此复杂的任务提供了最佳的准确性,速度和成本平衡.
科学领域:
- 计算化学计算化学
- 频谱学是一种光谱学.
- 机器学习 机器学习
背景情况:
- 核磁共振 (NMR) 光谱对于分子结构的确定至关重要.
- 由于自动化分析,NMR数据的数量越来越大,在结构性特征化方面造成了瓶.
- 手动解释NMR光谱是耗时且容易出现错误的.
研究的目的:
- 研究机器学习 (ML) 方法的潜力,特别是神经网络,以自动化NMR光谱解释.
- 为了将NMR光谱特征与分子亚结构相关联.
- 评估不同的神经网络架构和分子表示,用于NMR数据分析.
主要方法:
- 探索了三个神经网络架构:多层感知器 (MLP) +长短期记忆 (LSTM),卷积神经网络 (CNN) 和MLP +循环神经网络 (RNN).
- 分子表示包括功能组和基于邻居的新方法.
- 模型被训练在实验C和HNMR光谱上,有或没有实验元数据 (场强度,温度,溶剂).
主要成果:
- 在纳入实验元数据时,MLP + LSTM模型在C NMR光谱上实现了88%的准确性,比没有元数据的77%显著改善.
- CNN模型的准确性略低 (86%),但运行速度是MLP+LSTM模型的三倍.
- 鉴于准确性,计算时间和成本,CNN模型被确定为最实用的选择.
结论:
- 机器学习,特别是神经网络,可以有效地自动解释NMR光谱.
- 纳入实验元数据可以提高ML模型对NMR分析的准确性.
- 基于NMR的结构性特征的瓶,CNN模型提供了一个高效和准确的解决方案.
更多相关视频
相关概念视频
01:25Applications Of NMR In Biology
3.9K
Nuclear magnetic resonance (NMR) spectroscopy is a very valuable analytical technique for researchers. It has been used for more than 50 years as an analytical tool. F. Bloch and E. Purcell formulated NMR in 1946 and won the 1952 Nobel Prize in Physics for their work. Biological macromolecules such as proteins, nucleic acids, lipids, and organic molecules including pharmaceutical compounds, can be studied using this versatile tool that exploits the magnetic properties of certain nuclei.
3.9K
01:02¹H NMR: Interpreting Distorted and Overlapping Signals
1.1K
Spin systems where the difference in chemical shifts of the coupled nuclei is greater than ten times J are called first-order spin systems. These nuclei are weakly coupled, and their chemical shifts and coupling constant can generally be estimated from the well-separated signals in the spectrum.
As Δν decreases and the signals move closer, the doublets appear increasingly distorted. The intensities of the inner lines increase at the cost of those of the outer lines as the signals are...
As Δν decreases and the signals move closer, the doublets appear increasingly distorted. The intensities of the inner lines increase at the cost of those of the outer lines as the signals are...
1.1K
01:182D NMR: Overview of Heteronuclear Correlation Techniques
305
Heteronuclear correlation spectroscopy is an analytical technique that investigates the coupling between different types of nuclei, often a proton and an X-nucleus, such as carbon-13 or nitrogen-15. This method is commonly used in nuclear magnetic resonance (NMR) spectroscopy to gain insights into complex chemical compounds' structural and compositional aspects. A typical heteronuclear correlation spectrum displays X-nucleus chemical shifts on one axis and a proton spectrum on the other...
305
01:07Nuclear Magnetic Resonance (NMR): Overview
3.3K
Nuclear magnetic resonance (NMR) is a phenomenon exhibited by certain nuclei that can absorb characteristic radio frequency radiation under certain conditions. NMR has been extensively applied in molecular spectroscopy and medical diagnostic imaging. In both these applications, the molecule or subject under study is placed in a magnetic field and irradiated with radio frequency energy.
NMR spectroscopy generates a spectrum where the characteristic absorption frequencies of the sample are...
NMR spectroscopy generates a spectrum where the characteristic absorption frequencies of the sample are...
3.3K
01:12Two-Dimensional (2D) NMR: Overview
873
The 1D NMR spectrum of large and complex molecules like natural products has complicated splitting patterns and overlapping signals, which can be easily interpreted using 2-dimensional (2D) NMR. Unlike 1D NMR, 2D NMR has two frequency axes that provide the coupling information between the nucleus A and nucleus B in a molecule. The process from which 2D spectra are obtained has four steps.
The first step is the preparation period, during which nucleus A is excited with a radiofrequency pulse....
The first step is the preparation period, during which nucleus A is excited with a radiofrequency pulse....
873
01:14NMR Spectroscopy of Aromatic Compounds
5.0K
Aromatic compounds can be identified or analyzed using proton NMR and carbon‐13 NMR. Typically, aromatic hydrogens or hydrogens directly bonded to the aromatic rings are strongly deshielded by the aromatic ring current. Therefore, they absorb in the range of 6.5–8.0 ppm in proton NMR spectra. For instance, aromatic hydrogens directly bonded to the benzene ring absorb at 7.3 ppm. However, aromatic hydrogens of larger rings absorb farther upfield or downfield than the ideal range.
5.0K