acde (version 1.2.0)

qval: Q-Values Computation

Description

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

Usage

qval(Q, psi2)

Arguments

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.

Value

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

References

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 Unpublished MS thesis.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.

See Also

stp, fdr, ac2.

Examples

## 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"))