IsoTestBH {IsoGene} | R Documentation |
The function adjusts for the raw p-values of the five test statistics using BH or BY procedure.
IsoTestBH(rp, FDR, type = c("BH", "BY"), stat = c("E2", "Williams", "Marcus", "M", "ModifM"))
rp |
raw p-value matrix with each row for one gene and 6 columns, the first column contains the Probe.ID, the second to the sixth columns are raw p-values for the five test statistics |
FDR |
the desired FDR to control |
type |
choose BH or BY procedure to control FDR |
stat |
choose one of the five test statistics to use |
The input raw p-values to this function can be the one sided or the two sided ones which are obtained using function raw.p. The results using one sided p-values and FDR controlling at alpha/2 is equivalent to that using two sided p-values and FDR controlling at alpha.
sign.genes |
A list of significant genes while controlling FDR is obtained, with 4 columns: the first column is the probe ID, the second column is the row id, the third column is the raw p-values of the significant genes and the last column is the adjusted p-values of significant genes using BH or BY procedure |
This function only allows one type of FDR adjustment, either BH or
BY. For other type of adjustment, see function mt.rawp2adjp
in package multtest
.
Lin et al.
'mt.rawp2adjp', IsoRawp
set.seed(1234) rp <- data.frame(paste("g",1:100), matrix(runif(500,0,0.1), 100, 5)) sign <- IsoTestBH(rp, FDR = 0.05, type = "BH", stat = "E2")