Modelling highly co-eluted peaks of analytes with high spectral similarity

Ana Bordagaray; Jose Manuel Amigo Rubio

doi:10.1016/j.trac.2015.02.010

Modelling highly co-eluted peaks of analytes with high spectral similarity

Ana Bordagaray, Jose Manuel Amigo Rubio

Food Analytics and Biotechnology

7 Citationer (Scopus)

Abstract

Modelling co-eluted peaks has always been a keystone in chemometric applications to chromatographic data. Multivariate Curve Resolution (MCR) and Parallel Factor Analysis 2 (PARAFAC2) have been widely used to solve the co-elution problem (among others), profiting from the uniqueness of the spectrum for each co-eluted analyte. Nevertheless, there are cases where these curve-resolution approaches may fail, when the same analyte gives two co-eluted peaks or when two analytes have highly correlated spectra. In this article, we put forward the usefulness of Parallel Factor Analysis with Linear Dependence (PARALIND) that is able to handle overlapping peaks when the spectral profiles of the analytes are highly correlated. To illustrate the problem, we study two different situations: first, the co-elution of two peaks, belonging to L-proline; and, second, an overlapped mixture of fungicides with high similarity in their spectra.

Originalsprog	Engelsk
Tidsskrift	Trends in Analytical Chemistry
Vol/bind	68
Sider (fra-til)	107–118
Antal sider	12
ISSN	0165-9936
DOI	https://doi.org/10.1016/j.trac.2015.02.010
Status	Udgivet - 1 maj 2015

Adgang til dokumentet

10.1016/j.trac.2015.02.010

Citationsformater

@article{fdeaf984ef274276b73ce1c70c46496d,

title = "Modelling highly co-eluted peaks of analytes with high spectral similarity",

abstract = "Modelling co-eluted peaks has always been a keystone in chemometric applications to chromatographic data. Multivariate Curve Resolution (MCR) and Parallel Factor Analysis 2 (PARAFAC2) have been widely used to solve the co-elution problem (among others), profiting from the uniqueness of the spectrum for each co-eluted analyte. Nevertheless, there are cases where these curve-resolution approaches may fail, when the same analyte gives two co-eluted peaks or when two analytes have highly correlated spectra. In this article, we put forward the usefulness of Parallel Factor Analysis with Linear Dependence (PARALIND) that is able to handle overlapping peaks when the spectral profiles of the analytes are highly correlated. To illustrate the problem, we study two different situations: first, the co-elution of two peaks, belonging to L-proline; and, second, an overlapped mixture of fungicides with high similarity in their spectra.",

author = "Ana Bordagaray and {Amigo Rubio}, {Jose Manuel}",

year = "2015",

month = may,

day = "1",

doi = "10.1016/j.trac.2015.02.010",

language = "English",

volume = "68",

pages = "107–118",

journal = "Trends in Analytical Chemistry",

issn = "0165-9936",

publisher = "Elsevier",

}

TY - JOUR

T1 - Modelling highly co-eluted peaks of analytes with high spectral similarity

AU - Bordagaray, Ana

AU - Amigo Rubio, Jose Manuel

PY - 2015/5/1

Y1 - 2015/5/1

N2 - Modelling co-eluted peaks has always been a keystone in chemometric applications to chromatographic data. Multivariate Curve Resolution (MCR) and Parallel Factor Analysis 2 (PARAFAC2) have been widely used to solve the co-elution problem (among others), profiting from the uniqueness of the spectrum for each co-eluted analyte. Nevertheless, there are cases where these curve-resolution approaches may fail, when the same analyte gives two co-eluted peaks or when two analytes have highly correlated spectra. In this article, we put forward the usefulness of Parallel Factor Analysis with Linear Dependence (PARALIND) that is able to handle overlapping peaks when the spectral profiles of the analytes are highly correlated. To illustrate the problem, we study two different situations: first, the co-elution of two peaks, belonging to L-proline; and, second, an overlapped mixture of fungicides with high similarity in their spectra.

AB - Modelling co-eluted peaks has always been a keystone in chemometric applications to chromatographic data. Multivariate Curve Resolution (MCR) and Parallel Factor Analysis 2 (PARAFAC2) have been widely used to solve the co-elution problem (among others), profiting from the uniqueness of the spectrum for each co-eluted analyte. Nevertheless, there are cases where these curve-resolution approaches may fail, when the same analyte gives two co-eluted peaks or when two analytes have highly correlated spectra. In this article, we put forward the usefulness of Parallel Factor Analysis with Linear Dependence (PARALIND) that is able to handle overlapping peaks when the spectral profiles of the analytes are highly correlated. To illustrate the problem, we study two different situations: first, the co-elution of two peaks, belonging to L-proline; and, second, an overlapped mixture of fungicides with high similarity in their spectra.

U2 - 10.1016/j.trac.2015.02.010

DO - 10.1016/j.trac.2015.02.010

M3 - Journal article

SN - 0165-9936

VL - 68

SP - 107

EP - 118

JO - Trends in Analytical Chemistry

JF - Trends in Analytical Chemistry

ER -

Modelling highly co-eluted peaks of analytes with high spectral similarity

Abstract

Adgang til dokumentet

Fingeraftryk

Citationsformater