Effect of initial estimates and constraints selection in multivariate curve resolution—Alternating least squares. Application to low-resolution NMR data

Vivó-Truyols, Gabriel; Ziari, Maya; Magusin, Pieter C. M. M.; Schoenmakers, Peter J.

doi:10.1016/j.aca.2009.03.020

Cited by 6 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MCR‐ALS and orthogonal‐projection approach were applied to the series of Hahn‐echo 1 H NMR spectra of a cross‐linked unsaturated polyester resin . This is one of a few examples, where MCR was used to decompose solid state NMR data.…”

Section: Resultsmentioning

confidence: 99%

Independent component analysis (ICA) algorithms for improved spectral deconvolution of overlapped signals in ¹H NMR analysis: application to foods and related products

Monakhova

Tsikin

Kuballa

et al. 2014

Magnetic Reson in Chemistry

View full text Add to dashboard Cite

The major challenge facing NMR spectroscopic mixture analysis is the overlapping of signals and the arising impossibility to easily recover the structures for identification of the individual components and to integrate separated signals for quantification. In this paper, various independent component analysis (ICA) algorithms [mutual information least dependent component analysis (MILCA); stochastic non-negative ICA (SNICA); joint approximate diagonalization of eigenmatrices (JADE); and robust, accurate, direct ICA algorithm (RADICAL)] as well as deconvolution methods [simple-to-use-interactive self-modeling mixture analysis (SIMPLISMA) and multivariate curve resolution-alternating least squares (MCR-ALS)] are applied for simultaneous (1)H NMR spectroscopic determination of organic substances in complex mixtures. Among others, we studied constituents of the following matrices: honey, soft drinks, and liquids used in electronic cigarettes. Good quality spectral resolution of up to eight-component mixtures was achieved (correlation coefficients between resolved and experimental spectra were not less than 0.90). In general, the relative errors in the recovered concentrations were below 12%. SIMPLISMA and MILCA algorithms were found to be preferable for NMR spectra deconvolution and showed similar performance. The proposed method was used for analysis of authentic samples. The resolved ICA concentrations match well with the results of reference gas chromatography-mass spectrometry as well as the MCR-ALS algorithm used for comparison. ICA deconvolution considerably improves the application range of direct NMR spectroscopy for analysis of complex mixtures.

show abstract

Section: Resultsmentioning

confidence: 99%

Independent component analysis (ICA) algorithms for improved spectral deconvolution of overlapped signals in ¹H NMR analysis: application to foods and related products

Monakhova

Tsikin

Kuballa

et al. 2014

Magnetic Reson in Chemistry

View full text Add to dashboard Cite

show abstract

“…However, using competitive binding still requires the application of a T2 filter to suppress the protein background signal, at the expense of introducing T2-dependent attenuations in the "visible" LMWM signals, as indicated above. In this sense, the use of an additional NMR dimension based on differences in T2 relaxation (e.g., by acquiring a series of CPMG or Hahn spin−echo experiments with increasing tau delays) and multivariate curve resolution (MCR) allows separating components by their different T2 relaxation times 23 and provides T2 relaxation decays, from which T2corrected concentrations can be derived.…”

Section: ■ Introductionmentioning

confidence: 99%

Unravelling and Quantifying the “NMR-Invisible” Metabolites Interacting with Human Serum Albumin by Binding Competition and T2 Relaxation-Based Decomposition Analysis

et al. 2017

View full text Add to dashboard Cite

Quantitative profiling of low-molecular-weight metabolites (LMWMs) by H NMR is routinely used in high-throughput serum metabolomics. First, the protein background is attenuated using a T2 filter; then, the LMWM signals are resolved by line-shape fitting. However, protein-binding modifies the motional properties of LMWM, and their signal partially attenuates with the T2 filter, along with the protein background. Consequently, the quantified LMWM signals do not reflect the total concentration in serum but the nonbinding part. Here we present a novel strategy based on binding competition to promote the release of the "NMR-invisible" metabolites from serum proteins and achieve quantifications closer to total concentrations. The study focuses on five clinically relevant amino acids with different binding properties (valine, isoleucine, leucine, tyrosine, and phenylalanine). We analyzed their binding affinity to human serum albumin (HSA) in serum mimic samples and promoted the release of their bound fraction by TSP titration. Furthermore, we used a novel combination of pseudo-2D CPMG and multivariate curve resolution analysis, allowing the separation of LMWM and protein signals and providing LMWM quantifications corrected for transverse relaxation effects. We found that TSP concentrations larger than 3 mM released most of the bound fraction and validated these findings in real serum/plasma samples.

show abstract

“…The MCR-ALS method, known as a self-modeling algorithm, has been developed to decompose a data set into pure spectral responses and pure concentration profiles of all constituents present in unknown mixtures. So far, MCR-ALS has been exploited in many kinds of spectroscopy such as Raman spectroscopy, − near-infrared spectroscopy, − FT-IR, − UV–vis spectroscopy, − NMR spectroscopy, − high performance liquid chromatography, − gas chromatography, − and many others. At the end of the 1980s, a predecessor of the MCR-ALS algorithm (called SPFAC) was already used to explore EPR data sets. , Recently, the MCR-ALS method has been, successfully, applied on electron paramagnetic resonance spectroscopy .…”

mentioning

confidence: 99%

Extraction of Pure Spectral Signatures and Corresponding Chemical Maps from EPR Imaging Data Sets: Identifying Defects on a CaF₂ Surface Due to a Laser Beam Exposure

et al. 2015

View full text Add to dashboard Cite

A calcium fluoride (CaF2) plate was exposed to pulsed laser irradiations inducing surface morphological and ionization changes on its surface. More precisely surface damages mainly correspond to intrinsic defects. Electron paramagnetic resonance (EPR) hyperspectral imaging is a powerful technique able to characterize the defects formed on the CaF2 surface. Indeed, EPR hyperspectral images provide spatial and spectral information about the sample studied. In fact, these images possess a great potential to obtain accurate and reliable knowledge about the chemical composition and the distribution of the component due to the presence of the spatial aspect. However, the complexity of such hyperspectral data sets imposes the use of advanced chemometric tools to extract valuable information on the considered physicochemical system. Therefore, Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) is proposed to identify and locate the different constituents in the images. The originality of this work is that it reports on the application of MCR-ALS, for the first time, on electron paramagnetic resonance (EPR) imaging data sets that will furnish the distribution maps and the spectral signatures of all components present in the sample. The results show the identification of different intrinsic defects on a CaF2 sample from the sole information in the raw image measurements and, therefore, confirm the potential of this methodology and the important role of spatial information contained in the image.

show abstract

Effect of initial estimates and constraints selection in multivariate curve resolution—Alternating least squares. Application to low-resolution NMR data

Cited by 6 publications

References 25 publications

Independent component analysis (ICA) algorithms for improved spectral deconvolution of overlapped signals in ¹H NMR analysis: application to foods and related products

Independent component analysis (ICA) algorithms for improved spectral deconvolution of overlapped signals in ¹H NMR analysis: application to foods and related products

Unravelling and Quantifying the “NMR-Invisible” Metabolites Interacting with Human Serum Albumin by Binding Competition and T2 Relaxation-Based Decomposition Analysis

Extraction of Pure Spectral Signatures and Corresponding Chemical Maps from EPR Imaging Data Sets: Identifying Defects on a CaF₂ Surface Due to a Laser Beam Exposure

Contact Info

Product

Resources

About

Effect of initial estimates and constraints selection in multivariate curve resolution—Alternating least squares. Application to low-resolution NMR data

Cited by 6 publications

References 25 publications

Independent component analysis (ICA) algorithms for improved spectral deconvolution of overlapped signals in 1H NMR analysis: application to foods and related products

Independent component analysis (ICA) algorithms for improved spectral deconvolution of overlapped signals in 1H NMR analysis: application to foods and related products

Unravelling and Quantifying the “NMR-Invisible” Metabolites Interacting with Human Serum Albumin by Binding Competition and T2 Relaxation-Based Decomposition Analysis

Extraction of Pure Spectral Signatures and Corresponding Chemical Maps from EPR Imaging Data Sets: Identifying Defects on a CaF2 Surface Due to a Laser Beam Exposure

Contact Info

Product

Resources

About

Independent component analysis (ICA) algorithms for improved spectral deconvolution of overlapped signals in ¹H NMR analysis: application to foods and related products

Independent component analysis (ICA) algorithms for improved spectral deconvolution of overlapped signals in ¹H NMR analysis: application to foods and related products

Extraction of Pure Spectral Signatures and Corresponding Chemical Maps from EPR Imaging Data Sets: Identifying Defects on a CaF₂ Surface Due to a Laser Beam Exposure