NMR spectrum reconstruction as a pattern recognition problem

TL;DR

This paper introduces a WaveNet-based deep neural network that leverages pattern recognition to improve the reconstruction of NMR spectra from non-uniform sampling data, demonstrating superior results in simulations and real protein spectra.

Contribution

The novel WaveNet neural network architecture effectively utilizes prior knowledge of sampling patterns to enhance NMR spectrum reconstruction beyond existing methods.

Findings

Achieves high-quality reconstruction of 2D NMR spectra.

Successfully applied to spectra of proteins of various sizes.

Enables virtual homo-decoupling in methyl 1H-13 spectra.

Abstract

A new deep neural network based on the WaveNet architecture (WNN) is presented, which is designed to grasp specific patterns in the NMR spectra. When trained at a fixed non-uniform sampling (NUS) schedule, the WNN benefits from pattern recognition of the corresponding point spread function (PSF) pattern produced by each spectral peak resulting in the highest quality and robust reconstruction of the NUS spectra as demonstrated in simulations and exemplified in this work on 2D 1H-15N correlation spectra of three representative globular proteins with different sizes: Ubiquitin (8.6 kDa), Azurin (14 kDa), and Malt1 (44 kDa). The pattern recognition by WNN is also demonstrated for successful virtual homo-decoupling in a 2D methyl 1H-13 HMQC spectrum of MALT1. We demonstrate using WNN that prior knowledge about the NUS schedule, which so far was not fully exploited, can be used for designing new powerful NMR processing techniques that surpass the existing algorithmic methods.