rmRNAseq (version 0.1.0)

pauc_out: Evaluation of Differential Expression Analysis Methods

Description

This function calculates FDR (false discovery rate), pauc (partial area under ROC curve), when we know the vector of p-values obtained from a particular differential expression analysis method and the true status of each gene. The function requires that the first EE genes are true EE, and the last DE genes are true DE. This requirement can be fulfilled by reorder the rows of gene expression data set.

Usage

pauc_out(p, EE, DE)

Arguments

p

a vector of p-values.

EE

number of EE genes (the first EE genes in p-value vector p).

DE

number of DE genes (the last DE genes in p-value vector p).

Value

a vector including V (the number of false positives), R (the number of declared positives), FDR (the false discovery rate), pauc (the partial area under ROC curve with respect to false positive rate fpr less than or equal to a specified level), and auc. Here we consider 3 fpr: 0.05, 0.10, 0.20. So the output includes these 15 elements and the total auc.

Examples

# NOT RUN {
set.seed(1)
EE <- 1000
DE <- 500
p1 <- runif(EE)
p2 <- rbeta(DE, shape1 = .5, shape2 = 1)
p <- c(p1, p2)
rmRNAseq:::pauc_out(p, EE, DE)
# }

Run the code above in your browser using DataCamp Workspace