RNAither (version 2.20.0)

hitselectionZscorePval: Selecting hits according to ZScores and p-values

Description

Selects significant genes according to their ZScore (summarized with the gene median) and p-values.

Usage

hitselectionZscorePval(dataset, pValVec, col4zscore, col4sel, thresh, thresh2, 
flag2, col4anno, sumFunc, file4hits)

Arguments

dataset

an R data frame generated with generateDatasetFile

pValVec

a vector of p-values, as generated by one of the test functions Ttest, MannWhitney or RankProduct

col4zscore

a character vector specifying the name of the column containing the ZScores, usually "SigIntensity"

col4sel

a character vector specifying the name of the new dataset column where hits will be stored

thresh

the threshold for the ZScores. The interpretation depends on the choice of the parameter flag2.

thresh2

the threshold for the p-values

flag2

2 or -2. If 2 is chosen, all Zscores greater than or equal to thresh are chosen. If -2 is chosen, all Zscores smaller than or equal to thresh are chosen.

col4anno

a character string specifying the name of the dataset column to be used to define the replicate, for example "GeneName" or "Internal_GeneID".

sumFunc

the function used to summarize ZScore values, e.g. mean or median.

file4hits

the name of the file to store the results in

Value

A list containing:

dataset

the dataset with an added column defining the hits in the form of a binary vector

hitVector

the binary vector itself

thresh

the threshold for the ZScores

thresh2

he threshold for the p-values

ZScores and p-values are stored in a text output file.

Details

If there are no p-values under the defined threshold thresh2, it is increased to min(pvalvec). If flag2 == -2 and there are no ZScores under the defined threshold thresh, it is increased to min(ZScores). If flag2 == 2 and there are no ZScores over the defined threshold thresh, it is increased to max(ZScores). If there are not hits for the combined threshold of p-values and ZScores, the ZScore threshold is changed until there is a hit.

See Also

hitselectionPval, hitselectionZscore, Ttest

Examples

data(scoredDataset1, package="RNAither")
data(pValVec1, package="RNAither")

##for details on the generation of pValVec1 and scoredDataset1, see the example of the Ttest function linked above.

scoredHits1 <- hitselectionZscorePval(scoredDataset1, pValVec1, "SigIntensity", 
"Zscore_pval_hits", -1.5, 0.05, -2, "GeneName", median, "Zscores_pvals_testfile1.txt")

newdataset <- scoredHits1[[1]]
hitvector <- scoredHits1[[2]]