Centering, scaling, and transformations: improving the biological information content of metabolomics data

Berg, Robert A. van den; Hoefsloot, Huub C. J.; Westerhuis, Johan A.; Smilde, Age K.; Werf, M.J. van der

doi:10.1186/1471-2164-7-142

Cited by 1,997 publications

(1,414 citation statements)

References 28 publications

Supporting

Mentioning

1,394

Contrasting

Unclassified

Order By: Relevance

“…Peak areas were mean centered, level scaled to their means ( Van den Berg et al, 2006) and used to build a Partial Least Square Discriminant Analysis Model (PLS-DA), using the Classification Toolbox for MATLAB (Ballabio and Consonni, 2013). PLS-DA models (Positive ions/Negative ions) were built using the following parameters: two components, Bayes assignment, six cross validation (CV) groups in contiguous blocks (i.e., LeaveOne-Out validation as six samples were used).…”

Section: Untargeted Lipid Analysismentioning

confidence: 99%

Targeted and untargeted lipidomics of Emiliania huxleyi viral infection and life cycle phases highlights molecular biomarkers of infection, susceptibility, and ploidy

Hunter¹,

Frada²,

Fredricks³

et al. 2015

Front. Mar. Sci.

View full text Add to dashboard Cite

Marine viruses that infect phytoplankton strongly influence the ecology and evolution of their hosts. Emiliania huxleyi is characterized by a biphasic life cycle composed of a diploid (2N) and haploid (1N) phase; diploid cells are susceptible to infection by specific coccolithoviruses, yet haploid cells are resistant. Glycosphingolipids (GSLs) play a role during infection, but their molecular distribution in haploid cells is unknown. We present mass spectrometric analyses of lipids from cultures of uninfected diploid, infected diploid, and uninfected haploid E. huxleyi. Known viral GSLs were present in the infected diploid cultures as expected, but surprisingly, trace amounts of viral GSLs were also detected in the uninfected haploid cells. Sialic-acid GSLs have been linked to viral susceptibility in diploid cells, but were found to be absent in the haploid cultures, suggesting a mechanism of haploid resistance to infection. Additional untargeted high-resolution mass spectrometry data processed via multivariate analysis unveiled a number of novel biomarkers of infected, non-infected, and haploid cells. These data expand our understanding on the dynamics of lipid metabolism during E. huxleyi host/virus interactions and highlight potential novel biomarkers for infection, susceptibility, and ploidy.

show abstract

Section: Untargeted Lipid Analysismentioning

confidence: 99%

Targeted and untargeted lipidomics of Emiliania huxleyi viral infection and life cycle phases highlights molecular biomarkers of infection, susceptibility, and ploidy

Hunter¹,

Frada²,

Fredricks³

et al. 2015

Front. Mar. Sci.

View full text Add to dashboard Cite

show abstract

“…31 Commonly used methods of normalization in mass spectrometry and their influence on the identification of biomarkers in a non-image setting have been described elsewhere. [32][33][34][35] In the present study, seven methods of normalizing the spectral intensities of each pixel were evaluated. The factor by which the peak intensities of each individual pixel would be divided for each of these normalization methods is shown in Figure 3, where "informative peaks" refers to peaks that remained after variable selection with approaches A and B. Histogram matching is the only method that does not involve the division by a scalar normalization factor, and could therefore not be displayed.…”

Section: Normalizationmentioning

confidence: 99%

Robust Data Processing and Normalization Strategy for MALDI Mass Spectrometric Imaging

et al. 2012

View full text Add to dashboard Cite

Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) provides localized information about the molecular content of a tissue sample. To derive reliable conclusions about MSI data, it is necessary to implement appropriate processing steps in order to compare peak intensities across the different pixels comprising the image. Here, we review commonly used normalization methods, and propose a rational data processing strategy, for robust evaluation and modeling of MSI data. The approach includes newly developed heuristic methods for selecting biologically relevant peaks and pixels to reduce the size of a data set and remove the influence of the applied MALDI matrix. The methods are demonstrated on a MALDI MSI data set of a sagittal section of rat brain (4750 bins, m/z = 50-1000, 111 × 185 pixels) and the proposed preferred normalization method uses the median intensity of selected peaks, which were determined to be independent of the MALDI matrix. This was found to effectively compensate for a range of known limitations associated with the MALDI process and irregularities in MS image sampling routines. This new approach is relevant for processing of all MALDI MSI data sets, and thus likely to have impact in biomarker profiling, preclinical drug distribution studies, and studies addressing underlying molecular mechanisms of tissue pathology.

show abstract

“…Centering and scaling: centering and scaling pretreatments were applied to the data set by means of algorithms coded by authors (to more information about scaling preprocessing, lector is remitted to van der Berg et al 12 and literature cited herein).…”

Section: Resultsmentioning

confidence: 99%

“…Several preprocessing methods, including mean centering, autoscaling, range scaling, pareto scaling, vast scaling and level scaling, were applied to the dataset according to literature 12,13 . Before the scaling and centering processes a peak alignment was tested.…”

Section: Multivariate Analysismentioning

confidence: 99%

FINGERPRINT ANALYSIS OF UNFRACTIONATED Piper PLANT EXTRACTS BY HPLC-UV-DAD COUPLED WITH CHEMOMETRIC METHODS

Bernal

Delgado

Cuca

2012

J. Chil. Chem. Soc.

View full text Add to dashboard Cite

ABSTRACT23 crude ethanolic extracts of Piper plants were analyzed by combining HPLC fingerprints and chemometric methods including principal component analysis (PCA), hierarchical clustering analysis (HCA) and parallel factor analysis (PARAFAC). All statistical analysis were carried out both with and without previous preprocessing by counting with alignment of the fingerprints by correlation optimized warping (COW) and several normalization methods. Normalization processes of autoscaling, range scaling and vast scaling resulted to be the better offering the most relevant information. Both direct two-way methods (PCA and HCA) and HCA after data minimization employing a three-way method (PARAFAC) were demonstrated to be useful in the establishment of chemical composition relationships among Piper samples, and resulting in agreement with phylogenic relationships previously reported in literature to some Piper species.

show abstract

Centering, scaling, and transformations: improving the biological information content of metabolomics data

Cited by 1,997 publications

References 28 publications

Targeted and untargeted lipidomics of Emiliania huxleyi viral infection and life cycle phases highlights molecular biomarkers of infection, susceptibility, and ploidy

Targeted and untargeted lipidomics of Emiliania huxleyi viral infection and life cycle phases highlights molecular biomarkers of infection, susceptibility, and ploidy

Robust Data Processing and Normalization Strategy for MALDI Mass Spectrometric Imaging

FINGERPRINT ANALYSIS OF UNFRACTIONATED Piper PLANT EXTRACTS BY HPLC-UV-DAD COUPLED WITH CHEMOMETRIC METHODS

Contact Info

Product

Resources

About