Question

I’m having a problem with the function p.adjust. I have a list containing 741 p-values and I want to use the p.adjust() function to correct for multiple testing (FDR testing). This is what I have so far:

> x <- as.vector(pvalues1)
> p.adjust(x, method="fdr" n=length(x))

But I get the following error

Error in order (p, decreasing = TRUE) : 
     unimplemented type 'list' in 'orderVector1'

Can anyone help me with this?

Answer 1

The problem you have is the your list containing the p-values is a vector already. What you wanted was a numeric vector. A list is just a general vector:

> l <- list(A = runif(1), B = runif(1))
> l
$A
[1] 0.7053136

$B
[1] 0.7053284

> as.vector(l)
$A
[1] 0.7053136

$B
[1] 0.7053284

> is.vector(l)
[1] TRUE

One option is to unlist() the list, to produce a numeric vector:

> unlist(l)
        A         B 
0.7053136 0.7053284

the benefit of that is that it preserves the names. An alternative is plain old as.numeric(), which looses the names, but is otherwise the same as unlist():

> as.numeric(l)
[1] 0.7053136 0.7053284

For big vectors, you might not want to use the names in unlist(), so an alternative that will speed that version up is:

> unlist(l, use.names = FALSE)
[1] 0.7053136 0.7053284