Independent components analysis to increase efficiency of discriminant analysis methods (FDA and LDA): Application to NMR fingerprinting of wine

Monakhova, Yulia B.; Godelmann, Rolf; Kuballa, Thomas; Mushtakova, S. P.; Rutledge, Douglas N.

doi:10.1016/j.talanta.2015.03.037

Cited by 33 publications

(23 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Monakhova et al. have demonstrated the superiority of standard ICA over PCA and even over LDA, FDA, SIMCA, and PLS‐DA . The adaptation of ICA presented here orients the extraction of the most interesting ICs, without modifying their structure.…”

Section: Methodsmentioning

confidence: 81%

Urinary Metabolomics Profiles Associated to Bovine Meat Ingestion in Humans

Khodorova

Rutledge

Oberli

et al. 2018

Molecular Nutrition Food Res

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Methodsmentioning

confidence: 81%

Urinary Metabolomics Profiles Associated to Bovine Meat Ingestion in Humans

Khodorova

Rutledge

Oberli

et al. 2018

Molecular Nutrition Food Res

Self Cite

View full text Add to dashboard Cite

show abstract

“…Compared to PCA, ICA as a linear method could provide potential benefits for untargeted metabolomics. ICA has been successfully used in metabolomics (Li et al 2012 ; Monakhova et al 2015 ; Liu et al 2016 ). Other unsupervised methods, such as clustering, aim to identify naturally occurring clusters in the data set by using similarity measures defined by distance and linkage metrics (Wiwie et al 2015 ).…”

Section: Discussionmentioning

confidence: 99%

Advances in metabolome information retrieval: turning chemistry into biology. Part II: biological information recovery

Tebani

Afonso

Bekri

2017

J of Inher Metab Disea

View full text Add to dashboard Cite

This work reports the second part of a review intending to give the state of the art of major metabolic phenotyping strategies. It particularly deals with inherent advantages and limits regarding data analysis issues and biological information retrieval tools along with translational challenges. This Part starts with introducing the main data preprocessing strategies of the different metabolomics data. Then, it describes the main data analysis techniques including univariate and multivariate aspects. It also addresses the challenges related to metabolite annotation and characterization. Finally, functional analysis including pathway and network strategies are discussed. The last section of this review is devoted to practical considerations and current challenges and pathways to bring metabolomics into clinical environments.

show abstract

“…Among the dimensionality reduction and classification methods, the most popular ones include Principal Component Analysis (PCA) [164], Linear Discriminant Analysis (LDA) [165], Quadratic Discriminant Analysis (QDA) [166], Support Vector Machine (SVM) [167], Cluster Analysis (CA) [168], Factorial Discriminant Analysis (FDA) [169], Canonical Discriminant Analysis (CDA) [170], Hierarchical Clustering (HC) [171], and Artificial Neural Network (ANN) [120,164,172]. In turn, the concentration of samples is usually determined with Partial Least Squares (PLS) [173], Multiple Linear Regression (MLR) [174], Ridge Regression (RR), or regression ANN.…”

Section: Analysis Of Received Signalsmentioning

confidence: 99%

Methods for Early Detection of Microbiological Infestation of Buildings Based on Gas Sensor Technologies

et al. 2020

View full text Add to dashboard Cite

In this review, the problem of microbiological infestation of buildings was discussed. The techniques of detection were described as well, with special attention drawn to the rapid-early detection methods based on gas sensor arrays. The physical and chemical conditions of the building environment conducive to the development of microorganisms and the technical conditions influencing the problem of microbiological infestation were investigated. Additionally, the harmful effects on human health caused by the microbiological contamination were discussed, with a short review of particular groups of microorganisms causing sick building syndrome. Among the detection techniques, the traditional microbiological techniques as well as the molecular and chemical methods were presented. Different designs of the gas sensor arrays together with the various techniques of analyzing the received multidimensional signal were described, analyzed, and compared in detail.

show abstract

Independent components analysis to increase efficiency of discriminant analysis methods (FDA and LDA): Application to NMR fingerprinting of wine

Cited by 33 publications

References 55 publications

Urinary Metabolomics Profiles Associated to Bovine Meat Ingestion in Humans

Urinary Metabolomics Profiles Associated to Bovine Meat Ingestion in Humans

Advances in metabolome information retrieval: turning chemistry into biology. Part II: biological information recovery

Methods for Early Detection of Microbiological Infestation of Buildings Based on Gas Sensor Technologies

Contact Info

Product

Resources

About