View source: R/glog_transformation.R
glog_transformation  R Documentation 
Performs glog transformation on the data set. QC samples can be used to
estimate technical variation in the data set and calculate transformation
parameter \lambda
(lambda). QC samples usually comprise a pool of
aliquots taken from all other samples in the study and analysed repeatedly
throughout an analytical batch.
glog_transformation(df, classes, qc_label, lambda = NULL)
df 
A matrixlike (e.g. an ordinary matrix, a data frame) or
RangedSummarizedExperimentclass object with
all values of class 
classes 

qc_label 

lambda 

Many univariate and multivariate statistical tests require homogeneity and
n ormality of dataset variance. Realworld metabolomics datasets often fail
to meet these criteria due to asymmetric (i.e. non'normal') and/or
heteroscedatic (i.e. nonhomogenous) variance structure. To address this
issue, glog
data transformations may be applied.
For each cell within the data matrix, transform the raw value (x) according
to: log10(x + sqrt(x^2 + \lambda))
. The parameter \lambda
is
typically calculated using quality control (QC) samples analysed throughout
an analysis batch.
Object of class SummarizedExperiment
. If input data are a
matrixlike (e.g. an ordinary matrix, a data frame) object, function returns
the same R data structure as input with all value of data type
numeric()
.
Parsons HM et. al., BMC Bionf., 8(234), 2007. https://doi.org/10.1186/147121058234
df < MTBLS79[, MTBLS79$Batch == 1]
out < mv_imputation(df=df, method="knn")
out < glog_transformation (df=out, classes=df$Class,
qc_label="QC")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.