[R] p-adjust using Benjamn and Hochberg

[David Winsemius]

Keizer_71 <christophe.lo at gmail.com> wrote in
news:15944523.post at talk.nabble.com: 

> 
> Hello,
> 
> I am trying to use the p.adjust function for multiple testing.
> 
> here is what i have
> 
> 9997                201674_s_at  0.327547396
> 9998                221013_s_at  0.834211067
> 9999                221685_s_at  0.185099475
> 
> I import them from excel have have the gene symbol as well as the
> pvalue 
> 
> here is the issue
> 
>> pa<-p.adjust(pt,method="BH")
> Error in p[nna] : object is not subsettable
> In addition: Warning message:
> In is.na(p) : is.na() applied to non-(list or vector) of type
> 'closure' 
>> adjust<-sum(pa<0.1)
> 
> Is there anything i need to change?

Two that I can think of:

A) Change your posting style. Start including reproducible code. You 
are making the audience guess at what pt might be. At the very least 
you could have included the results of str(pt).

B) Change the argument to p.adjust so that it is the vector of p-values 
rather than what is (probably) the entire dataframe.

See:

gt<-"9997 201674_s_at  0.327547396
9998      221013_s_at  0.834211067
9999      221685_s_at  0.185099475
"
gene.p<-read.table(file=textConnection(gt),header=FALSE)
pa<-p.adjust(gene.p$V3,method="BH")

> pa
[1] 0.4913211 0.8342111 0.4913211

-- 
David Winsemius