Premium
Resolving NMR signals of short‐chain fatty acid mixtures using unsupervised component analysis
Author(s) -
Costa Pereira Jorge,
Jarak Ivana,
Carvalho Rui Albuquerque
Publication year - 2017
Publication title -
magnetic resonance in chemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.483
H-Index - 72
eISSN - 1097-458X
pISSN - 0749-1581
DOI - 10.1002/mrc.4606
Subject(s) - chemistry , deconvolution , biological system , component (thermodynamics) , multivariate statistics , independent component analysis , nuclear magnetic resonance spectroscopy , characterization (materials science) , signal (programming language) , proton nmr , nmr spectra database , multivariate analysis , analytical chemistry (journal) , pattern recognition (psychology) , chromatography , artificial intelligence , spectral line , nanotechnology , computer science , machine learning , organic chemistry , algorithm , physics , materials science , astronomy , biology , programming language , thermodynamics
Nuclear magnetic resonance (NMR) is a very powerful instrumental technique suited to identify and characterize organic compounds. NMR has been successfully used in the analysis of complex biological and environmental samples; however, these applications are still rather limited. In this work, we describe unsupervised component analysis as a multivariate unsupervised method suited to identify the number of relevant NMR signal contributions and to deconvolve mixed signals into signal individual sources and respective contributions. Using this approach, we were able to advance further in the field of quantification of NMR spectra, and this methodology will help in the characterization of complex biological samples. Copyright © 2017 John Wiley & Sons, Ltd.