Standard screening for numeric traits based on Pearson correlation.
standardScreeningNumericTrait(datExpr, yNumeric, corFnc = cor,
corOptions = list(use = 'p'),
alternative = c("two.sided", "less", "greater"),
qValues = TRUE,
areaUnderROC = TRUE)
Data frame with the following components:
Gene (or variable) identifiers copied from colnames(datExpr)
correlations of all genes with the trait
Fisher Z statistics corresponding to the correlations
Student p-values of the correlations
(if input qValues==TRUE) q-values of the correlations calculated from the p-values
(if input areaUnderROC==TRUE) area under the ROC
number of samples present for the calculation of each association.
data frame containing expression data (or more generally variables to be screened), with rows corresponding to samples and columns to genes (variables)
a numeric vector giving the trait measurements for each sample
correlation function.
Defaults to Pearson correlation but can also be bicor.
list specifying additional arguments to be passed to the correlation function given
by corFnc.
alternative hypothesis for the correlation test
logical: should q-values be calculated?
logical: should are under the receiver-operating curve be calculated?
Steve Horvath
The function calculates the correlations, associated p-values, area under the ROC, and q-values
standardScreeningBinaryTrait, standardScreeningCensoredTime