groupChromPeaks-nearest: Peak grouping based on proximity in the mz-rt space
In yclement/xcms: LC-MS and GC-MS Data Analysis

Description Usage Arguments Value Slots Note References See Also Examples

This method is inspired by the grouping algorithm of mzMine (Katajamaa 2006) and performs correspondence based on proximity of peaks in the space spanned by retention time and mz values. The method creates first a master peak list consisting of all chromatographic peaks from the sample in which most peaks were identified, and starting from that, calculates distances to peaks from the sample with the next most number of peaks. If peaks are closer than the defined threshold they are grouped together.

The NearestPeaksParam class allows to specify all settings for the peak grouping based on the nearest algorithm. Instances should be created with the NearestPeaksParam constructor.

sampleGroups,sampleGroups<-: getter and setter for the sampleGroups slot of the object.

mzVsRtBalance,mzVsRtBalance<-: getter and setter for the mzVsRtBalance slot of the object.

absMz,absMz<-: getter and setter for the absMz slot of the object.

absRt,absRt<-: getter and setter for the absRt slot of the object.

kNN,kNN<-: getter and setter for the kNN slot of the object.

groupChromPeaks,XCMSnExp,NearestPeaksParam: performs peak grouping based on the proximity between chromatographic peaks from different samples in the mz-rt range.

The correspondence analysis can be performed on chromatographic peaks of any MS level (if present and if chromatographic peak detection has been performed for that MS level) defining features combining these peaks. The MS level can be selected with the parameter msLevel. By default, calling groupChromPeaks will remove any previous correspondence results. This can be disabled with add = TRUE, which will add newly defined features to already present feature definitions.

NearestPeaksParam(
  sampleGroups = numeric(),
  mzVsRtBalance = 10,
  absMz = 0.2,
  absRt = 15,
  kNN = 10
)

## S4 method for signature 'NearestPeaksParam'
show(object)

## S4 method for signature 'NearestPeaksParam'
sampleGroups(object)

## S4 replacement method for signature 'NearestPeaksParam'
sampleGroups(object) <- value

## S4 method for signature 'NearestPeaksParam'
mzVsRtBalance(object)

## S4 replacement method for signature 'NearestPeaksParam'
mzVsRtBalance(object) <- value

## S4 method for signature 'NearestPeaksParam'
absMz(object)

## S4 replacement method for signature 'NearestPeaksParam'
absMz(object) <- value

## S4 method for signature 'NearestPeaksParam'
absRt(object)

## S4 replacement method for signature 'NearestPeaksParam'
absRt(object) <- value

## S4 method for signature 'NearestPeaksParam'
kNN(object)

## S4 replacement method for signature 'NearestPeaksParam'
kNN(object) <- value

## S4 method for signature 'XCMSnExp,NearestPeaksParam'
groupChromPeaks(object, param, msLevel = 1L, add = FALSE)

`sampleGroups`	A vector of the same length than samples defining the sample group assignments (i.e. which samples belong to which sample group). This parameter is mandatory for the `PeakDensityParam` and has to be provided also if there is no sample grouping in the experiment (in which case all samples should be assigned to the same group).
`mzVsRtBalance`	`numeric(1)` representing the factor by which mz values are multiplied before calculating the (euclician) distance between two peaks.
`absMz`	`numeric(1)` maximum tolerated distance for mz values.
`absRt`	`numeric(1)` maximum tolerated distance for rt values.
`kNN`	`numeric(1)` representing the number of nearest neighbors to check.
`object`	For `groupChromPeaks`: an XCMSnExp object containing the results from a previous chromatographic peak detection analysis (see `findChromPeaks()`). For all other methods: a `NearestPeaksParam` object.
`value`	The value for the slot.
`param`	A `PeakDensityParam` object containing all settings for the peak grouping algorithm.
`msLevel`	`integer(1)` defining the MS level. Currently only MS level 1 is supported.
`add`	`logical(1)` (default `add = FALSE`) allowing to perform an additional round of correspondence (e.g. on a different MS level) and add features to the already present feature definitions.

The NearestPeaksParam function returns a NearestPeaksParam class instance with all of the settings specified for peak alignment based on peak proximity.

For groupChromPeaks: a XCMSnExp object with the results of the peak grouping/correspondence step (i.e. the mz-rt features). These can be accessed with the featureDefinitions() method.

.__classVersion__,sampleGroups,mzVsRtBalance,absMz,absRt,kNN: See corresponding parameter above. .__classVersion__ stores the version from the class. Slots values should exclusively be accessed via the corresponding getter and setter methods listed above.

These methods and classes are part of the updated and modernized xcms user interface. All of the settings to the algorithm can be passed with a NearestPeaksParam object.

Katajamaa M, Miettinen J, Oresic M: MZmine: Toolbox for processing and visualization of mass spectrometry based molecular profile data. Bioinformatics 2006, 22:634-636.

The do_groupChromPeaks_nearest() core API function.

featureDefinitions() and featureValues() for methods to access peak grouping results (i.e. the features).

XCMSnExp for the object containing the results of the peak grouping.

Other peak grouping methods: groupChromPeaks-density, groupChromPeaks-mzClust, groupChromPeaks()

## Create a NearestPeaksParam object
p <- NearestPeaksParam(kNN = 3)
p

##############################
## Chromatographic peak detection and grouping.
##
## Below we perform first a chromatographic peak detection (using the
## matchedFilter method) on some of the test files from the faahKO package
## followed by a peaks grouping using the "nearest" method.
library(faahKO)
library(MSnbase)
fls <- dir(system.file("cdf/KO", package = "faahKO"), recursive = TRUE,
           full.names = TRUE)

## Reading 2 of the KO samples
raw_data <- readMSData(fls[1:2], mode = "onDisk")

## Perform the peak detection using the matchedFilter method.
mfp <- MatchedFilterParam(snthresh = 20, binSize = 1)
res <- findChromPeaks(raw_data, param = mfp)

head(chromPeaks(res))
## The number of peaks identified per sample:
table(chromPeaks(res)[, "sample"])

## Performing the peak grouping
p <- NearestPeaksParam()
res <- groupChromPeaks(res, param = p)

## The results from the peak grouping:
featureDefinitions(res)

## Using the featureValues method to extract a matrix with the intensities of
## the features per sample.
head(featureValues(res, value = "into"))

## The process history:
processHistory(res)