Description Usage Arguments Details Value S3 METHODS References See Also Examples
Creates a data quality object using a set of spectra from a quality control (QC) sample. The result can be used to assess the quality of other spectra generated from the same QC sample.
1 |
x |
A matrix of peak intensity values with spectra as rows and
peak classes as columns. The peak intensity matrix can be
estimated via the |
... |
Additional arguments for the specified principal component analysis
|
FUN |
A character string specifying the method for principal component analysis.
Possible choices are |
The user is expected to provide a (training) peak intensity matrix
that has been derived from a set of pooled quality control samples.
The output of msQualify
contains the
projection of this matrix onto its principal
components (PCs) via the princomp
or princompRob
function.
The user can subsequently
assess the quality of another (test) peak intensity matrix generated from the
same QC sample via the predict
method, which compares the training PCs
to the test PCs.
An object of class msQualify
.
Predict the quality of a set of spectra. This method supports the following optional arguments.
An object of class msQualify
.
A matrix of peak intensities.
It must have the same number of columns as the peak intensity matrix
used to compute the msQualify
object.
A character string indicating the criterion to be use. Possible choices are "Cattell" and "Kaiser". Default: "Cattell".
A numeric value representing the threshold to be used. Default: 0.9.
Coombes KR, Fritsche HA Jr., Clarke C, Chen JN, Baggerly KA, Morris JS, Xiao LC, Hung MC, and Kuerer HM, “Quality control and peak finding for proteomics data collected from nipple aspirate fluid by surface-enhanced laser desorption and ionization," Clinical Chemistry, 49(10), pp. 1615–23, 2003.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 | ## create multiple reference samples with multiple
## peaks
set.seed(10)
nrs <- 240
nv <- 35
my.mean <- 10
my.sd <- rnorm(nv)
my.sd <- my.sd - min(my.sd) + 1
rsam <- splus2R::rmvnorm(n=nrs, d=nv, mean=rep(my.mean, nv),
cov=diag(nv), sd=my.sd)
## run msQualify
pca <- msQualify(rsam, FUN="princompRob", estim="auto")
## create multiple reference samples with multiple
## peaks from the same distribution
nts <- 72
tsam <- splus2R::rmvnorm(n=nts, d=nv, mean=rep(my.mean, nv),
cov=diag(nv), sd=my.sd)
## predict the quality of the test samples
quality <- predict(pca, tsam)
quality$pass
if (!is.R()) assign("quality", quality, frame=1)
## check if the distances truly follow
## chisq(nkeep) distribution
qqmath(~quality$dist,
distribution=function(p, df=quality$df) qchisq(p, df),
panel = function(x, y) {
panel.grid()
panel.abline(0, 1)
panel.qqmath(x, y)
},
aspect=1,
xlab=paste("Chisq(", quality$df, ") Quantile"),
ylab="mahalanobis distance")
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.