“Ask Ernö”: a self-learning tool for assignment and prediction of nuclear magnetic resonance spectra

Machine learning (ML) methods have been present in the field of NMR since decades, but it has experienced a tremendous growth in the last few years, especially thanks to the emergence of deep learning (DL) techniques taking advantage of the increased amounts of data and available computer power. These algorithms are successfully employed for classification, regression, clustering, or dimensionality reduction tasks of large data sets and have been intensively applied in different areas of NMR including metabonomics, clinical diagnosis, or relaxometry. In this article, we concentrate on the various applications of ML/DL in the areas of NMR signal processing and analysis of small molecules, including automatic structure verification and prediction of NMR observables in solution. K E Y W O R D S

Section: Nmr Predictionmentioning

confidence: 99%

NMR signal processing, prediction, and structure verification with machine learning techniques

Cobas¹

2020

“…But beyond the importance of testing sets to validate new methods for NMR analysis is the requirement of larger datasets allowing the method itself to function. Chemical shift prediction is the clearest example: All state‐of‐the‐art chemical shift predictors need a database of assigned spectra to work . For all we know, this is probably the way it will always be .…”

Section: Discussionmentioning

confidence: 99%

“…The first repositories of raw spectra appeared at least a decade ago . Their importance is increasingly recognized for various applications and strategies, such as fingerprinting; computer‐assisted spectra analysis; design of QSAR/QSPR descriptors to predict properties from spectra; and identification of putative metabolites . Nevertheless, it appears that availability of raw NMR data is still a significant issue.…”

Section: Introductionmentioning

confidence: 99%

The C6H6 NMR repository: An integral solution to control the flow of your data from the magnet to the public

Patiny

Zasso

Kostro

et al. 2017

Self Cite

NMR is a mature technique that is well established and adopted in a wide range of research facilities from laboratories to hospitals. This accounts for large amounts of valuable experimental data that may be readily exported into a standard and open format. Yet the publication of these data faces an important issue: Raw data are not made available; instead, the information is slimed down into a string of characters (the list of peaks). Although historical limitations of technology explain this practice, it is not acceptable in the era of Internet. The idea of modernizing the strategy for sharing NMR data is not new, and some repositories exist, but sharing raw data is still not an established practice. Here, we present a powerful toolbox built on recent technologies that runs inside the browser and provides a means to store, share, analyse, and interact with original NMR data. Stored spectra can be streamlined into the publication pipeline, to improve the revision process for instance. The set of tools is still basic but is intended to be extended. The project is open source under the Massachusetts Institute of Technology (MIT) licence.

“…This should require only small efforts (see Table ) by software developers to include ‘export’ tools and a bit of motivation on the side of the chemists to generate data. Adding the content of diverse existing databases, even if records are partial and include only 1D data, could quickly represent a reliable source of information for computer program able to take such data into account …”

Section: Advantages Of a Human‐ And Computer‐readable 2d Correlation mentioning

confidence: 99%

Human‐ and computer‐accessible 2D correlation data for a more reliable structure determination of organic compounds. Future roles of researchers, software developers, spectrometer managers, journal editors, reviewers, publisher and database managers toward artificial‐intelligence analysis of NMR spectra

Jeannerat

2016

The introduction of a universal data format to report the correlation data of 2D NMR spectra such as COSY, HSQC and HMBC spectra will have a large impact on the reliability of structure determination of small organic molecules. These lists of assigned cross peaks will bridge signals found in NMR 1D and 2D spectra and the assigned chemical structure. The record could be very compact, human and computer readable so that it can be included in the supplementary material of publications and easily transferred into databases of scientific literature and chemical compounds. The records will allow authors, reviewers and future users to test the consistency and, in favorable situations, the uniqueness of the assignment of the correlation data to the associated chemical structures. Ideally, the data format of the correlation data should include direct links to the NMR spectra to make it possible to validate their reliability and allow direct comparison of spectra. In order to take the full benefits of their potential, the correlation data and the NMR spectra should therefore follow any manuscript in the review process and be stored in open-access database after publication. Keeping all NMR spectra, correlation data and assigned structures together at all time will allow the future development of validation tools increasing the reliability of past and future NMR data. This will facilitate the development of artificial intelligence analysis of NMR spectra by providing a source of data than can be used efficiently because they have been validated or can be validated by future users. Copyright © 2016 John Wiley & Sons, Ltd.