R: Data Structure for multiple alignment of many GCMS samples

multipleAlignment-class {flagme}

R Documentation

Data Structure for multiple alignment of many GCMS samples

Description

Store the raw data and optionally, information regarding signal peaks for a number of GCMS runs

Usage

multipleAlignment(pd, group, bw.gap = 0.8, wn.gap = 0.6, bw.D = 0.20, 
                         wn.D = 0.05, filterMin = 1, lite = FALSE, usePeaks = TRUE,
                         df = 50, verbose = TRUE, timeAdjust = FALSE, 
                         doImpute = FALSE, metric = 2, type = 2, penality = 0.2)

Arguments

`pd`	a `peaksDataset` object
`group`	factor variable of experiment groups, used to guide the alignment algorithm
`bw.gap`	gap parameter for "between" alignments
`wn.gap`	gap parameter for "within" alignments
`bw.D`	distance penalty for "between" alignments
`wn.D`	distance penalty for "within" alignments
`filterMin`	minimum number of peaks within a merged peak to be kept in the analysis
`lite`	logical, whether to keep "between" alignment details (default, `FALSE`)
`usePeaks`	logical, whether to use peaks (if `TRUE`) or the full 2D profile alignment (if `FALSE`)
`df`	distance from diagonal to calculate similarity
`verbose`	logical, whether to print information
`timeAdjust`	logical, whether to use the full 2D profile data to estimate retention time drifts (Note: time required)
`doImpute`	logical, whether to impute the location of unmatched peaks
`metric`	numeric, different algorithm to calculate the similarity matrix between two mass spectrum. `metric=1` call `normDotProduct()`; `metric=2` call `ndpRT()`; `metric=3` call `corPrt()`
`type`	numeric, two different type of alignment function
`penality`	penalization applied to the matching between two mass spectra if `(t1-t2)>D`

Details

multipleAlignment is the data structure giving the result of an alignment across several GCMS runs.

Multiple alignments are done progressively. First, all samples with the same tg$Group label with be aligned (denoted a "within" alignment). Second, each group will be summarized into a pseudo-data set, essentially a spectrum and retention time for each matched peak of the within-alignment. Third, these "merged peaks" are aligned in the same progressive manner, here called a "between" alignment.

Value

multipleAlignment object

Author(s)

Mark Robinson

References

Mark D Robinson (2008). Methods for the analysis of gas chromatography - mass spectrometry data PhD dissertation University of Melbourne.

Examples

	require(gcspikelite)

	## paths and files
	gcmsPath <- paste(find.package("gcspikelite"), "data", sep = "/")
	cdfFiles <- dir(gcmsPath, "CDF", full = TRUE)
	eluFiles <- dir(gcmsPath, "ELU", full = TRUE)

	## read data, peak detection results
	pd <- peaksDataset(cdfFiles[1:2], mz = seq(50, 550), rtrange = c(7.5, 8.5))
	pd <- addAMDISPeaks(pd,eluFiles[1:2])

	## multiple alignment
	ma <- multipleAlignment(pd, c(1, 1), wn.gap = 0.5, wn.D = 0.05, bw.gap = 0.6,
		                    bw.D = 0.2, usePeaks = TRUE, filterMin = 1, df = 50,
		                    verbose = TRUE, metric = 1, type = 1)

[Package flagme version 1.43.0 Index]