qval: Q-Values Computation
In acde: Artificial Components Detection of Differentially Expressed Genes

Description Usage Arguments Value Author(s) References See Also Examples

For internal use in function stp. Computes the genes' Q-Values in the Single Time Point Analysis according to Algorithm 3 in the vignette.

1	qval(Q, psi2)

`Q`	vector with the estimated FDRs when the threshold values used are `abs(ac2(Z, design))`.
`psi2`	vector with the second artificial component as returned by `ac2`.

returns a vector with the computed Q-Values for each gene in the experiment.

Juan Pablo Acosta (jpacostar@unal.edu.co).

Acosta, J. P. (2015) Strategy for Multivariate Identification of Differentially Expressed Genes in Microarray Data. Unpublished MS thesis. Universidad Nacional de Colombia, Bogot\'a.

Storey, J. D. (2002) A direct approach to false discovery rates. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 64(3): 479–498.

stp, fdr, ac2.

## Single time point analysis for 500 genes with 10 treatment 
## replicates and 10 control replicates
n <- 500; p <- 20; p1 <- 10
des <- c(rep(1, p1), rep(2, (p-p1)))
mu <- as.matrix(rexp(n, rate=1))
Z <- t(apply(mu, 1, function(mui) rnorm(p, mean=mui, sd=1)))
### 5 up regulated genes
Z[1:5,1:p1] <- Z[1:5,1:p1] + 5
### 10 down regulated genes
Z[6:15,(p1+1):p] <- Z[6:15,(p1+1):p] + 4

res <- fdr(Z, des)
qValues <- qval(res$Q, ac2(Z, des))
plot(res$th, res$Q, type="l", col="blue")
lines(res$th, qValues[order(abs(ac2(Z, des)))], col="green")
legend(x="topright", legend=c("FDR", "Q Values"), lty=c(1,1), 
    col=c("blue", "green"))