Bucketing: Spectral data reduction
In ManonMartin/SOAP-NMR: Pre-process 1H-NMR FID signals

Description Usage Arguments Details Value Author(s) References Examples

Reduces the number of data points by aggregating intensities into buckets.

1 2	Bucketing(Spectrum_data, width = FALSE, mb = 500, boundary = NULL, intmeth = c("r", "t"), tolbuck = 10^-4, verbose = FALSE)

`Spectrum_data`	Matrix containing the spectra in ppm, one row per spectrum.
`width`	If `width` is `TRUE`, then `m` represents the buckets width, otherwise, it represents the number of buckets.
`mb`	The number of buckets OR the buckets' width. If `mb` represents the number of buckets, it should be an integer smaller or equal to the number of frequencies in `Spectrum_data`.
`boundary`	Numeric vector of left and right boundaries for ppm integration.
`intmeth`	Type of bucketing: rectangular (`"r"`) or trapezoidal (`"t"`). See details below.
`tolbuck`	Tolerance threshold to check if the buckets of the original spectra are of constant length.
`verbose`	If`"TRUE"`, will print processing information.

It is important to note that the input spectrum can have its ppm axis in increasing or decreasing order and it does not have to be equispaced.

Bucketing has two main interests:

Ease the statistical analysis
Decrease the impact of peaks misalignments between different spectra that should be aligned; assuming we are in the ideal case where they fall in the same bucket. Of course, the better the prior warping is, the larger m can be without major misalignment and the more informative the spectra will be.

The ppm interval of Spectrum_data, let's say [a,b] where a > b, is divided into mb buckets of size (a-b)/mb. The new ppm scale contains the m centers of these intervals. The spectral intensity at these centers is the integral of the initial spectral intensity on this bucket using either trapezoidal or rectangular integration.

Spectrum_data

The matrix of spectra with their new ppm axis.

Benoît Legat, Bernadette Govaerts & Manon Martin

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).