... Neither siggenes nor samr can correctly handle technical replicates, as far as I know. If you used the classlabel as you describe then you will consider technical replicates as if they were biological replicates, which all things equal should bias you towards the alternative hypothesis. In other wor ...
... Running summary on a list is probably not that great an idea, as you doubtless found out. Instead, it's preferable to first find the class of an object and then query from there.
... The MBNI custom cdfs are generated by a group that is completely distinct from Bioconductor. If you have questions about their cdfs, please ask them, not us.
... The probeset IDs for the MBNI custom cdfs are the database ID concatenated to _st. So e.g., if you were using the Entrez Gene version, given an Entrez Gene ID of 1223, the probeset ID would be 1223_st.
If we assume you have run rma to get an ExpressionSet (that you called 'eset'), you can then make ...
... It's just a matter of coercion, isn't it?
> myGenes <- data.frame(matrix(c(1572, 1571, 4129, NA, 216, 217, 15, 3620, 7453,NA), 2))
X1 X2 X3 X4 X5
1 1572 4129 216 15 7453
2 1571 NA 217 3620 NA
> myGenes2 <- data.frame(matrix(mapIds(org.Hs.eg.db, as.charact ...
... You can look at the GPL for the array that the submitter chose when submitting the microarray data, which is on the GEO page for the data you are using.
... You have a complete mismatch between your Bioconductor and R versions. BioC 3.2 was intended for R-3.2.x
You should restart R, then do
Which should get you the correct version of BiocInstaller, which will then install the correct versi ...
... It's not clear what you are trying to do. Why not just use the existing package for those data? Re-creating an already extant package is a waste of time, I would think.
... This comes from BiocGenerics, in particular the unsafe_replaceSlots function. The code is pretty well commented, so I will point you there for a rationale.