Metabolomic data preprocessing strategy for 1H-NMR spectroscopic data

Description

This package provides R functions for classic and advanced preprocessing steps that are applied on 1H NMR data. It also provides the function ReadFids to read the FID directly from the Bruker format. Those preprocessing are cited below in the advised order of their application:

GroupDelayCorrection

Correct for the first order phase correction.

SolventSuppression

Remove solvent signal from the FIDs.

Apodization

Increase the sensitivity/resolution of the FIDs.

ZeroFilling

Improve the visual visual representation of the spectra.

FourierTransform

Transform the FID into a spectrum and convert the frequency scale (Hertz -> ppm).

ZeroOrderPhaseCorrection

Correct for the zero order phase correction.

InternalReferencing

Calibrate the spectra with internal compound referencing.

BaselineCorrection

Remove the spectral baseline.

NegativeValuesZeroing

Set negatives values to 0.

Warping

Warp the samples according to a reference spectrum.

WindowSelection

Select the informative part of the spectrum.

Bucketing

Data reduction by integration.

RegionRemoval

Set intensities of a desired region to 0.

ZoneAggregation

Aggregate a region to a single peak.

Normalization

Normalize the spectra.

Details

The FIDs are read using ReadFids which also gives a matrix with meta-information about each FID. The other functions apply different preprocessing steps on these signals, and some need the info matrix as outputted from ReadFids . During this preprocessing, the signal is transformed through fourier transformation and the frequency scale is expressed in ppm. For more details and illustrated explanations about those pre-treatment steps, see the documentation of each function and/or the chapter 1 of the reference below.

Author(s)

Benoît Legat, Bernadette Govaerts & Manon Martin

Maintainer: Manon Martin <manon.martin@uclouvain.be>

References

Martin, M., Legat, B., Leenders, J., Vanwinsberghe, J., Rousseau, R., Boulanger, B., & Govaerts, B. (2018). PepsNMR for 1H NMR metabolomic data pre-processing. Analytica chimica acta, 1019, 1-13.

Rousseau, R. (2011). Statistical contribution to the analysis of metabonomics data in 1H NMR spectroscopy (Doctoral dissertation, PhD thesis. Institut de statistique, biostatistique et sciences actuarielles, Université catholique de Louvain, Belgium).

Examples

path <-  system.file("extdata", package = "PepsNMRData")
dir(path)

fidList <- ReadFids(file.path(path, "HumanSerum"))
Fid_data <- fidList[["Fid_data"]]
Fid_info <- fidList[["Fid_info"]]
Fid_data <- GroupDelayCorrection(Fid_data, Fid_info)
Fid_data <- SolventSuppression(Fid_data)
Fid_data <- Apodization(Fid_data, Fid_info)
Fid_data <- ZeroFilling(Fid_data)
Spectrum_data <- FourierTransform(Fid_data, Fid_info)
Spectrum_data <- ZeroOrderPhaseCorrection(Spectrum_data)
Spectrum_data <- InternalReferencing(Spectrum_data, Fid_info)
Spectrum_data <- BaselineCorrection(Spectrum_data)
Spectrum_data <- NegativeValuesZeroing(Spectrum_data)
Spectrum_data <- Warping(Spectrum_data)
Spectrum_data <- WindowSelection(Spectrum_data)
Spectrum_data <- Bucketing(Spectrum_data)
Spectrum_data <- RegionRemoval(Spectrum_data, typeofspectra = "serum")
# Spectrum_data <- ZoneAggregation(Spectrum_data)
Spectrum_data <- Normalization(Spectrum_data, type.norm = "mean")