[R] Multiple testing corrections on very large vector

[terdon]

Hi Jorge, and thanks for your answer, it looks promising.

However, I have a question. First of all, I am a lowly biologist so please
excuse any horrible errors of understanding I may make.

So, the BH correction depends on, among other things, sorting the vector of
P-values from the smallest to the largest and then multiplying each value by
the length of the vector times its rank.

If I understand correctly, your suggestion is to use  p.adjust on a subset
of the vector pv, set the n parameter to the length of the entire vector and
then iterate until all P-values have been adjusted.

 Wouldn't I still need to know the rank of each p value in the subset vector
with respect to its position in the original sorted vector? I can see how to
calculate that of course, but how can I give this info to p.adjust if it
only takes the length of the vector as an argument?

Thanks again

--
View this message in context: http://r.789695.n4.nabble.com/Multiple-testing-corrections-on-very-large-vector-tp3341398p3341615.html
Sent from the R help mailing list archive at Nabble.com.