mrbin (version 1.9.4)

annotatemrbin: A function for annotating mrbin objects.

Description

This function annotates an mrbin object and returns it with updated $annotations vector

Usage

annotatemrbin(
  mrbinObject,
  annotate = TRUE,
  binwidth = 0.01,
  binheight = 1,
  dimension = "1D",
  metaboliteIdentities = NULL,
  add = FALSE,
  hideChemicalShift = FALSE,
  tiers = 1:5,
  hideTentativeIds = TRUE,
  confirmationRthreshold = 0.7,
  confirmationPthreshold = 1e-05,
  uniqueBins = TRUE,
  checkBaselineCorrelation = FALSE,
  verbose = TRUE
)

Value

An (invisible) mrbin object

Arguments

mrbinObject

An mrbin object or matrix of bin intensities. In case of a matrix, the row names can be in the format left border, right border ("1.22,1.21") or the center of the bin ("1.215"). In the latter case, the parameters binwidth (and binheight if required) should be chosen accordingly. For 2D data, this would be "1.22,1.21,35,36" or "1.215,35.5"

annotate

If FALSE, the mrbin object will not be changed.

binwidth

Full width of each bin. Will only be used if mrbinObject is a matrix

binheight

Full height of each bin. Will only be used if mrbinObject is a matrix and dimension is set to "2D"

dimension

Dimension of NMR data set, option are "1D" or "2D" (e.g. for HSQC data). Will only be used if mrbinObject is a matrix

metaboliteIdentities

A numeric 4-column matrix or the file path for a .csv file containing such a matrix, the first columns containing metabolite names and the first row being a header. Each row belongs to one unique metabolite signal (left, right, top, bottom borders). Row names are metabolite names. If provided, this will overwrite any current metaboliteIdentities matrix present in the mrbin object. If missing, data currently attached to the mrbin object (if any) will be used.

add

Should the new metabolite list be added to an existing list, or replace the current list?

hideChemicalShift

Should the chemical shift (bin borders) of an identified metabolite be removed, leaving only the metabolite id, or should both be shown? Showing both helps in identifying signals of interest, but hiding the chemical shift might make better plots.

tiers

Should all tiers (1 through 5) of identification be applied? .

hideTentativeIds

Should the identities of tentative ids be omitted for clarity?

confirmationRthreshold

A threshold to define the r-value cutoff to confirm an annotation

confirmationPthreshold

A threshold to define the p-value cutoff to confirm an annotation

uniqueBins

Should each bin be uniquely assigned to only one molecule?

checkBaselineCorrelation

Should correlation to baseline be compared to confirm an annotation

verbose

Should outputs be shown or suppressed?

Examples

resetEnv()#clean up previous data from the package environment 
results<-mrbin(silent=TRUE,
                   parameters=list(verbose=TRUE,dimension="1D",PQNScaling="No",
                   binwidth1D=0.04,signal_to_noise1D=1,PCA="No",binRegion=c(9.5,0.5,10,156),
                   saveFiles="No",referenceScaling="No",noiseRemoval="No",
                   fixNegatives="No",logTrafo="No",noiseThreshold=.05,
                   NMRvendor="mrbin",
                   example=TRUE,#only used for the package examples
                   NMRfolders=c(system.file("extdata/3.mr1",package="mrbin"),
                              system.file("extdata/2.mr1",package="mrbin"),
                              system.file("extdata/1.mr1",package="mrbin"))))
metaboliteIdentities=matrix(c(1.346,1.324,21,23,1,1,
                              3.052,3.043,30.5,33.5,1,1),
                   ncol=6,byrow=TRUE)
rownames(metaboliteIdentities)=c("Lactate","Creatinine")
colnames(metaboliteIdentities)=c("left","right","top","bottom","usePeak1D","usePeak2D")
results<-annotatemrbin(results,metaboliteIdentities=metaboliteIdentities)
results$metadata$annotations[150:160]
plotPCA(results,loadings=TRUE)