A Bayesian approach for the alignment of high-resolution NMR spectra

Abstract

Metabolic analysis with high-resolution nuclear magnetic resonance (NMR) enables simultaneous investigation of numerous chemical species in response to biochemical changes in subjects. When the analysis involves comparing two or more NMR spectra, it is essential to properly align them because small variations across different spectra influence the alignment and thus, interfere with direct comparisons between samples. We propose a new alignment method within the Bayesian modeling framework. The proposed method allows us to estimate the amplitude and phase shifts simultaneously and to obtain robust results in the existence of noise. Effectiveness of our proposed method is demonstrated through real NMR spectra in human plasma and a comparison study with dynamic time warping and correlated optimized warping, two widely used alignment methods in spectral data.

Getting Your Peaks in Line: A Review of Alignment Methods for NMR Spectral Data

“…Besides stretching/compression and shifting, a third group of methods uses a polynomial model for correction. Eilers et al [ 18 ] and Kim et al [ 25 ] directly estimate the aligned spectra from their warping functions through polynomial models.…”

“…A first group of methods, considers the whole spectra for alignment. For example PTW [ 18 ] and BAA [ 25 ] both present each spectrum as a function of data points, build the models of the shifts between two spectra and minimize the difference between the spectra. FW [ 21 ] uses Fuzzy warping to find maximally corresponding peaks in whole spectra, and then uses an interpolation function for alignment.…”

“…We can visualize a number of relevant features in a few different ways, as illustrated in Figure 4 . A simple visualization are spectral plots [ 10 , 13 , 16 , 19 , 20 , 21 , 23 , 25 ] in which whole spectra or a region are overlaid and plotted to see how the spectra were aligned ( Figure 4 a). A related and effective approach consists of presenting all spectra in a single image of size N x M , where N is the number of spectra and M is the number of data points in each spectrum.…”

Metabolômica De Plantas: Métodos E Desafios

“…255,256 Dentre os algoritmos utilizados para correção dessas mudanças locais, destacam-se o Partial Linear Fit (PLF), 256 Peak Alignment by Genetic Algorithm (PAGA), 165 Peak Alignment using Reduced Set (PARS), 257 Peak Alignment by Beam Search (PABS), 258 Peak Alignment by PCA (PAPCA), 259 Fuzzy Warping (FW), 260 Generlized Fuzzy Hought Transform (GFHT), 261 Recursive Segment wise Peak Alignment (RSPA), 262 Progressive Consensus Alignment of NMR Spectra (PCANS), 263 Hierarchical Cluster-based Peak Alignment (CluPA), 264 e Bayesian approach for alignment (BAA). 265 Interval COrrelated Shifting (ICOShift) é a função de alinhamento mais utilizada em metabolômica para dados de RMN. 266 Nesse algoritmo, os espectros são divididos em diferentes segmentos e alinhados segundo um espectro de referência (média ou variância do conjunto amostral pode também ser utilizado).…”

Preprocessing Methods in Nuclear Magnetic Resonance Spectroscopy

“…This is done by adding or removing points in the spectrum. Other algorithms for matching peaks in the NMR spectra, such as fuzzy warping [42], Bayesian alignment [18] and the FFT peak alignment [40] apply a predetermined pattern, which is translated into a set of spectra.…”