A pilot study for fragment identification using 2D NMR and deep learning

Stefan Kuhn; Eda Tumer; Simon Colreavy-Donnelly; Ricardo Moreira Borges

doi:10.1002/mrc.5212

A pilot study for fragment identification using 2D NMR and deep learning

Stefan Kuhn, Eda Tumer, Simon Colreavy-Donnelly, Ricardo Moreira Borges

Research output: Contribution to journal › Article › peer-review

Abstract

This paper presents a proof of concept of a method to identify substructures in 2D NMR spectra of mixtures using a bespoke image-based convolutional neural network application. This is done using HSQC and HMBC spectra separately and in combination. The application can reliably detect substructures in pure compounds, using a simple network. Results indicate that it can work for mixtures when trained on pure compounds only. HMBC data and the combination of HMBC and HSQC show better results than HSQC alone in this pilot study.

Original language	English
Pages (from-to)	1052-1060
Number of pages	9
Journal	Magnetic Resonance in Chemistry
Volume	60
Issue number	11
DOIs	https://doi.org/10.1002/mrc.5212
Publication status	Published - Nov 2022
Externally published	Yes

Keywords

convolutional neural network
deep learning
image processing
NMR
structure elucidation

Access to Document

10.1002/mrc.5212

Cite this

@article{fa804e312bf34c988d7fad16483ead7b,

title = "A pilot study for fragment identification using 2D NMR and deep learning",

abstract = "This paper presents a proof of concept of a method to identify substructures in 2D NMR spectra of mixtures using a bespoke image-based convolutional neural network application. This is done using HSQC and HMBC spectra separately and in combination. The application can reliably detect substructures in pure compounds, using a simple network. Results indicate that it can work for mixtures when trained on pure compounds only. HMBC data and the combination of HMBC and HSQC show better results than HSQC alone in this pilot study.",

keywords = "convolutional neural network, deep learning, image processing, NMR, structure elucidation",

author = "Stefan Kuhn and Eda Tumer and Simon Colreavy-Donnelly and {Moreira Borges}, Ricardo",

note = "Publisher Copyright: {\textcopyright} 2021 John Wiley & Sons, Ltd.",

year = "2022",

month = nov,

doi = "10.1002/mrc.5212",

language = "English",

volume = "60",

pages = "1052--1060",

journal = "Magnetic Resonance in Chemistry",

issn = "0749-1581",

number = "11",

}

TY - JOUR

T1 - A pilot study for fragment identification using 2D NMR and deep learning

AU - Kuhn, Stefan

AU - Tumer, Eda

AU - Colreavy-Donnelly, Simon

AU - Moreira Borges, Ricardo

PY - 2022/11

Y1 - 2022/11

N2 - This paper presents a proof of concept of a method to identify substructures in 2D NMR spectra of mixtures using a bespoke image-based convolutional neural network application. This is done using HSQC and HMBC spectra separately and in combination. The application can reliably detect substructures in pure compounds, using a simple network. Results indicate that it can work for mixtures when trained on pure compounds only. HMBC data and the combination of HMBC and HSQC show better results than HSQC alone in this pilot study.

AB - This paper presents a proof of concept of a method to identify substructures in 2D NMR spectra of mixtures using a bespoke image-based convolutional neural network application. This is done using HSQC and HMBC spectra separately and in combination. The application can reliably detect substructures in pure compounds, using a simple network. Results indicate that it can work for mixtures when trained on pure compounds only. HMBC data and the combination of HMBC and HSQC show better results than HSQC alone in this pilot study.

KW - convolutional neural network

KW - deep learning

KW - image processing

KW - NMR

KW - structure elucidation

UR - http://www.scopus.com/inward/record.url?scp=85115213017&partnerID=8YFLogxK

U2 - 10.1002/mrc.5212

DO - 10.1002/mrc.5212

M3 - Article

C2 - 34480494

AN - SCOPUS:85115213017

SN - 0749-1581

VL - 60

SP - 1052

EP - 1060

JO - Magnetic Resonance in Chemistry

JF - Magnetic Resonance in Chemistry

IS - 11

ER -

A pilot study for fragment identification using 2D NMR and deep learning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this