wavSmoothing {MALDIrppa}

Discrete wavelet transformation for MassSpectrum objects

Description

This function performs undecimated wavelet transform (UDWT) on mass spectra in MassSpectrum format. Alternatively, smoothing methods included in the MALDIquant package can be called.

Usage

wavSmoothing(x, method = c("Wavelet", "SavitzkyGolay", "MovingAverage"),
             n.levels = 4, ...)

Arguments

x

A list of MassSpectrum objects.

method

Smoothing method used.

n.levels

Depth of the decomposition for wavelet-based smoothing.

...

Other arguments.

Details

Note that from version 1.1.0 of MALDIrppa wavelet smoothing is conducted by maximal overlap discrete wavelet transformation and universal thresholding of coefficients based on methods available on the waveslim package. The optimal level of smoothing is determined by model-driven estimates of the thresholds. The parameter n.levels (values > 0 and <= log(length(x),2)) can be used to tweak the levels to obtain a smoother or rougher result.

Alternatively, smoothing methods SavitzkyGolay and MovingAverage from the MALDIquant package can be called directly from this function.

If the previous implementation of the wavelet method is required please download and install manually source files of version 1.0.5-1 from the archive of old sources of the package (https://CRAN.R-project.org/package=MALDIrppa).

Value

A list of MassSpectrum objects with denoised signal intensities.

Examples

# Load example data

data(spectra) # list of MassSpectra class objects

# sqrt transformation and signal smoothing using UDWT

spectra <- transfIntensity(spectra, fun = "sqrt")
spectra <- wavSmoothing(spectra)