[R] Hi

[Francois Pepin]

Hi Jiang,

where did you get that definition of the Benjamini-Hochberg correction? That is simply not how it works. You can take a look at the original paper, it is available online (http://www.math.tau.ac.il/~ybenja/MyPapers/benjamini_hochberg1995.pdf). You can also look at the p.adjust function to see how it is implemented in R.

Also, please keep the replies on the list. They're archived so that other people can refer to it later it also gives other people the opportunity to reply if they have more to add.

Francois

On Sep 30, 2011, at 15:07 , chunjiang he wrote:

> Hi Francois,
>  
> Thansk for your reply. I did BH correction manually. Just use corrected pvalue=pvalue*(n/n-1). And I got the result like this: The pvalue in manual and by R package are different.
>  
> title	r	raw	R package	manual
> hsa-miR-205--GATA3	0.797883	1.08E-13	1.08E-12	1.08E-12
> hsa-miR-205--ITGB4	0.750218	1.85E-11	8.20E-11	9.25E-11
> hsa-miR-187--PGF	0.797604	3.24E-11	8.20E-11	1.08E-10
> hsa-miR-205--SERPINB5	0.744125	3.28E-11	8.20E-11	8.20E-11
> hsa-miR-205--PBX1	0.734487	7.89E-11	1.58E-10	1.58E-10
> hsa-miR-205--MCC	0.724999	1.80E-10	3.00E-10	3.00E-10
> hsa-miR-205--WNT5B	0.717705	3.33E-10	3.78E-10	4.76E-10
> hsa-miR-200c--PKN2	0.721747	3.46E-10	3.78E-10	4.33E-10
> hsa-miR-200c--PCYOX1	0.721698	3.48E-10	3.78E-10	3.87E-10
> hsa-miR-200c--WDR68	0.72068	3.78E-10	3.78E-10	3.78E-10
> 
> So I am confused about that.
>  
> Best,
> Jiang
> On Fri, Sep 30, 2011 at 4:53 PM, Francois Pepin <francois.pepin at sequentainc.com> wrote:
> Hi Jiang,
> 
> you'll have to give us some more information than this: a reproducible example and why you expect things to be different.
> 
> This method has been tested extensively so we'd need something more specific if we are to help you.
> 
> Francois
> 
> On Sep 30, 2011, at 14:40 , chunjiang he wrote:
> 
> > Hi,
> >
> > There is a question that I am confused.
> > I have a set of data like this:
> >
> > hsa-miR-205--GATA3    0.797882767 1.08E-13
> > hsa-miR-205--ITGB4     0.750217593 1.85E-11
> > hsa-miR-187--PGF        0.797604155 3.24E-11
> > hsa-miR-205--SERPINB5 0.744124886 3.28E-11
> > hsa-miR-205--PBX1 0.734487224 7.89E-11
> > hsa-miR-205--MCC 0.72499934 1.80E-10
> > hsa-miR-205--WNT5B 0.717705259 3.33E-10
> > hsa-miR-200c--PKN2 0.721746815 3.46E-10
> > hsa-miR-200c--PCYOX1 0.721698034 3.48E-10
> > hsa-miR-200c--WDR68 0.72068017 3.78E-10
> >
> > And I want to do the Benjamini & Hochberg correction.
> >
> > So I run :
> >
> > rm(list=ls())
> > a<-read.csv("1-correlation.txt",sep="\t",header=F,quote="")
> > c<-p.adjust(a$V3,"BH")
> > a[,4]<-c
> > write.table(a,"zz.txt",sep="\t")
> >
> > And I got the result:
> >
> > hsa-miR-205--GATA3 0.797882767 1.08E-13 1.08E-12
> > hsa-miR-205--ITGB4 0.750217593 1.85E-11 8.20E-11
> > hsa-miR-187--PGF 0.797604155 3.24E-11 8.20E-11
> > hsa-miR-205--SERPINB5 0.744124886 3.28E-11 8.20E-11
> > hsa-miR-205--PBX1 0.734487224 7.89E-11 1.58E-10
> > hsa-miR-205--MCC 0.72499934 1.80E-10 3.00E-10
> > hsa-miR-205--WNT5B 0.717705259 3.33E-10 3.78E-10
> > hsa-miR-200c--PKN2 0.721746815 3.46E-10 3.78E-10
> > hsa-miR-200c--PCYOX1 0.721698034 3.48E-10 3.78E-10
> > hsa-miR-200c--WDR68 0.72068017 3.78E-10 3.78E-10
> >
> > When I check it, I found some adjust pvalue are not same to result that I
> > did manually.
> >
> > Can anyone help for this. Thanks,
> >
> > Jiang
> >
> >       [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help at r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
> 
> 
>