Ultrafast diffusion-based unmixing of <sup>1</sup>H NMR spectra

Mishra, Rituraj; Marchand, Achille; Jacquemmoz, Corentin; Dumez, Jean‐Nicolas

doi:10.1039/d0cc07757g

Cited by 5 publications

(7 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For DECRA, this required the design of alternative frequency swept pulses, and resulted in a duration of just a few second for both the acquisition and the processing/analysis steps. 164 The combination of UF 2D NMR and single-shot hyperpolarisation methods is particularly promising for mixture analysis. 102,[165][166][167][168][169] This is notably the case for dissolution DNP, which provides high polarisation levels for a broad range of molecules, but which is also not generally compatible with multi-scan pulse sequences.…”

Section: Ultrafast 2d Nmrmentioning

confidence: 99%

See 1 more Smart Citation

NMR methods for the analysis of mixtures

Dumez

2022

Chem. Commun.

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Ultrafast 2d Nmrmentioning

confidence: 99%

“…For DECRA, this required the design of alternative frequency swept pulses, and resulted in a duration of just a few second for both the acquisition and the processing/analysis steps. 164 …”

Section: Changementioning

confidence: 99%

NMR methods for the analysis of mixtures

Dumez

2022

Chem. Commun.

Self Cite

View full text Add to dashboard Cite

show abstract

“…They are successfully used for the identification of pure substances and the verification of complex samples. In many cases, mixtures of small molecules are common in chemistry, and they can be analyzed by NMR spectroscopy directly without further separation and purification [36,37]. Due to the signal overlap and interferences in NMR spectra, the previously mentioned similarity methods may fail in analyzing the mixtures.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning-Based Method for Compound Identification in NMR Spectra of Mixtures

et al. 2022

View full text Add to dashboard Cite

Nuclear magnetic resonance (NMR) spectroscopy is highly unbiased and reproducible, which provides us a powerful tool to analyze mixtures consisting of small molecules. However, the compound identification in NMR spectra of mixtures is highly challenging because of chemical shift variations of the same compound in different mixtures and peak overlapping among molecules. Here, we present a pseudo-Siamese convolutional neural network method (pSCNN) to identify compounds in mixtures for NMR spectroscopy. A data augmentation method was implemented for the superposition of several NMR spectra sampled from a spectral database with random noises. The augmented dataset was split and used to train, validate and test the pSCNN model. Two experimental NMR datasets (flavor mixtures and additional flavor mixture) were acquired to benchmark its performance in real applications. The results show that the proposed method can achieve good performances in the augmented test set (ACC = 99.80%, TPR = 99.70% and FPR = 0.10%), the flavor mixtures dataset (ACC = 97.62%, TPR = 96.44% and FPR = 2.29%) and the additional flavor mixture dataset (ACC = 91.67%, TPR = 100.00% and FPR = 10.53%). We have demonstrated that the translational invariance of convolutional neural networks can solve the chemical shift variation problem in NMR spectra. In summary, pSCNN is an off-the-shelf method to identify compounds in mixtures for NMR spectroscopy because of its accuracy in compound identification and robustness to chemical shift variation.

show abstract

“…The essential smoothing features of CONTIN may broaden all the peaks. , Multivariate methods analyze the total information available in the data simultaneously. Antalek et al developed a direct exponential curve resolution algorithm (DECRA) based on the generalized rank annihilation method. , DECRA can resolve the spectra of components whose diffusion coefficients differ by less than 20% . Nevertheless, the approach is somewhat sensitive to noise .…”

Section: Introductionmentioning

confidence: 99%

“…23,24 DECRA can resolve the spectra of components whose diffusion coefficients differ by less than 20%. 25 Nevertheless, the approach is somewhat sensitive to noise. 18 Additionally, the Multivariate Curve Resolution (MCR) method is also a promising way of analyzing the data from DOSY experiments.…”

Section: ■ Introductionmentioning

confidence: 99%

Retrieving Spectra of Pure Components from the DOSY-NMR Experiment via a Comprehensive Approach Involving the 2D Asynchronous Spectrum, 2D Quotient Spectrum, and Genetic Algorithm Refinement

et al. 2022

Anal. Chem.

View full text Add to dashboard Cite

When diffusion coefficients of different components in a mixture are similar, NMR spectra of pure individual components are difficult to be obtained via a diffusion-ordered spectroscopy (DOSY) experiment. Two-dimensional correlation spectroscopy (2D-COS) is used to analyze the data from the DOSY experiment. Through the properties of the systematic absence of cross-peak (SACP) in the 2D asynchronous spectra, spectra of pure components can be obtained even if their diffusion coefficients are similar. However, fluctuations in peak-position and peak-width are often unavoidable in NMR spectra, which makes SACPs unrecognizable. To address the problem, a 2D quotient spectrum is used to identify the masked SACPs. However, undesirable interference peaks due to the fluctuations in peak-position and peak-width are still present when we extract a spectrum of a component by slicing the 2D asynchronous spectrum across the SACP. A genetic algorithm (GA) is used to select a suitable subset of spectra where the diversities of peak-position and peak-width are significantly reduced. Then, we used the selected spectra to construct a refined 2D asynchronous spectrum so that the spectra of pure components with significant attenuated interference can be obtained. The above approach has been proven to be effective on a model system and a real-world example, demonstrating that 2D-COS possesses a bright perspective in the analysis of the bilinear data from DOSY experiments.

show abstract

Ultrafast diffusion-based unmixing of ¹H NMR spectra

Abstract: We show that the NMR spectra of components in a mixture can be separated using 2D data acquired in less than one second, and an algorithm that is executed in...

Cited by 5 publications

References 30 publications

NMR methods for the analysis of mixtures

NMR methods for the analysis of mixtures

Deep Learning-Based Method for Compound Identification in NMR Spectra of Mixtures

Retrieving Spectra of Pure Components from the DOSY-NMR Experiment via a Comprehensive Approach Involving the 2D Asynchronous Spectrum, 2D Quotient Spectrum, and Genetic Algorithm Refinement

Contact Info

Product

Resources

About

Ultrafast diffusion-based unmixing of 1H NMR spectra

Abstract: We show that the NMR spectra of components in a mixture can be separated using 2D data acquired in less than one second, and an algorithm that is executed in...

Cited by 5 publications

References 30 publications

NMR methods for the analysis of mixtures

NMR methods for the analysis of mixtures

Deep Learning-Based Method for Compound Identification in NMR Spectra of Mixtures

Retrieving Spectra of Pure Components from the DOSY-NMR Experiment via a Comprehensive Approach Involving the 2D Asynchronous Spectrum, 2D Quotient Spectrum, and Genetic Algorithm Refinement

Contact Info

Product

Resources

About

Ultrafast diffusion-based unmixing of ¹H NMR spectra