Dear All,
i have performed a limma paired analysis on an affymetrix expression dataset(13 cancer vs 13 control samples).
My initial non-specific filtered expressionSet which was used as input for the limma analysis: data.trusted.eset
and my design matrix looks like this:
(Intercept) conditionCancer pairs2 pairs3 pairs4 pairs5 pairs6 pairs7 pairs8 pairs9 pairs10 pairs11.....
1 1 0 0 0 0 0 0 0 0 0 0 0
2 1 1 0 0 0 0 0 0 0 0 0 0
3 1 0 1 0 0 0 0 0 0 0 0 0
4 1 1 1 0 0 0 0 0 0 0 0 0
5 1 0 0 1 0 0 0 0 0 0 0 0
6 1 1 0 1 0 0 0 0 0 0 0 0
7 1 0 0 0 1 0 0 0 0 0 0 0
8 1 1 0 0 1 0 0 0 0 0 0 0
9 1 0 0 0 0 1 0 0 0 0 0 0
10 1 1 0 0 0 1 0 0 0 0 0 0
11 1 0 0 0 0 0 1 0 0 0 0 0
12 1 1 0 0 0 0 1 0 0
The specific coefficient im interested is "conditionCancer"
So i used the code below:
x <- hgu133aPATH2PROBE
mapped_probes <- mappedkeys(x)
xx <- as.list(x[mapped_probes])
indices <- ids2indices(xx, rownames(data.trusted.eset))
res <- mroast(data.trusted.eset, indices, design, contrast=2)
Firstly, my main question is if the implementation of the mroast fuction is correct regarding the argument contrast=2 in which i believe refers to the coefficient="conditionCancer" ?
Secondly, should i filter my data.trusted.eset with the output of the topTable function after removing possible duplicates and probesets mapping to NAs, in order to remove possible false results ?
For instance:
filtered <- data.trusted.eset[rownames(top3),] # where top3 is the final dataframe of topTable function after removing duplicates and NAs
and then use filtered instead of the initial expressionset ??
Dear kslowikowski, thank you for your verification about the argument contrast. Moreover, i often use the function treat() or rank the probesets(genes) based on various criteria, but i have never considered decideTests, so i will see how it works