1. Neural net analysis of NMR spectra from strongly-coupled spin systems.
- Author
-
Prestegard, James H., Boons, Geert-Jan, Chopra, Pradeep, Glushka, John, Grimes, John H., and Simon, Bernd
- Subjects
- *
COUPLING constants , *QUANTUM theory , *ARTIFICIAL intelligence , *INTEGRATED software , *SPECTRUM analysis - Abstract
[Display omitted] • The neural network described extracts coupling constants and chemical shifts from a 2D J-resolved spectrum. • Training of the network on spectra simulated using quantum dynamics is demonstrated. • Results on simulated test sets are satisfactory, but experimental spectra present obstacles. • Future work employing pure-phase J-resolved spectra is suggested. Extracting parameters such as chemical shifts and coupling constants from proton NMR spectra is often a first step in using spectra for compound identification and structure determination. This can become challenging when scalar couplings between protons are comparable in size to chemical shift differences (strongly coupled), as is often the case with low-field (bench top) spectrometers. Here we explore the potential utility of AI methods, in particular neural networks, for extracting parameters from low-field spectra. Rather than seeking large experimental sets of spectra for training a network, we chose quantum mechanical simulation of sets, something that is possible with modern software packages and computer resources. We show that application of a network trained on 2-D J-resolved spectra and applied to a spectrum of iduronic acid, shows some promise, but also meets with some obstacles. We suggest that these may be overcome with improved pulse sequences and more extensive simulations. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF