mzFilter-methods: Filter the features of an imaging dataset by intensity
In Cardinal: A mass spectrometry imaging toolbox for statistical analysis

Description Usage Arguments Details Value Author(s) See Also Examples

Apply filtering to a mass spectrometry imaging dataset based on the intensities of each peak or mass feature.

## S4 method for signature 'MSImagingExperiment'
mzFilter(object, ..., freq.min = NA, rm.zero = TRUE)

## S4 method for signature 'MSImagingExperiment'
peakFilter(object, ..., freq.min = 0.01, rm.zero = TRUE)

## S4 method for signature 'MSImageSet'
peakFilter(object, method = "freq", ...)

## Filter based on the frequency of a peak
peakFilter.freq(x, freq.min=0.01, ...)

`object`	An object of class `MSImageSet`.
`freq.min`	Minimum frequency; peaks that occur in the dataset in lesser proportion than this will be dropped.
`rm.zero`	Remove features with mean intensities of zero.
`...`	Additional arguments passed to the peak filtering method, or conditions evaluating to logical vectors where only those conditions that are TRUE are retained.
`method`	The peak filtering method to use.
`x`	The vector of ion image intensities to filter.

When applied to a MSImagingExperiment object, mzFilter and peakFilter uses the summarize() to generate useful summary statistics about the mass features or detected peaks. These include the ‘min’, ‘max’, ‘mean’, ‘sum’, ‘sd’, and ‘var’ of the intensities for each mass feature or peak. These can be used in logical expressions to filter the features of the dataset.

Note that peakFilter is an alias for mzFilter, with different default parameters that are more appropriate for peak-picked data rather than profile spectra.

When applied to a MSImageSet object, unlike most other processing methods, peakFilter operates on the feature space (ion images) of the dataset.

Peak filtering is usually performed using the provided functions, but a user-created function can also be passed to method. In this case it should take the following arguments:

x: The vector of ion image intensities to filter.
...: Additional arguments.

A user-created function should return a logical: TRUE means keep the peak, and FALSE means remove the peak.

Internally, featureApply is used to apply the filtering. See its documentation page for more details on additional objects available to the environment installed to the peak filtering function.

An object of the same class with the filtered peaks.

Kylie A. Bemis

MSImagingExperiment, MSImageSet, peakPick, peakAlign, peakBin, reduceDimension, featureApply, process

setCardinalBPPARAM(SerialParam())

set.seed(2)
data <- simulateImage(preset=1, npeaks=10, dim=c(3,3))
data <- data[,pData(data)$circle]

# filter m/z features
process(mzFilter(data))

# queue peak picking, alignment, and filtering
data <- peakPick(data, method="simple", SNR=6)
data <- peakAlign(data, tolerance=200, units="ppm")
data <- peakFilter(data, freq.min=0.5)

# apply peak picking, alignment, and filtering
data_peaks <- process(data, plot=interactive())