[BioC] ? bug in limma classifyTestsP

[Gordon Smyth]

At 03:53 AM 7/08/2004, Kimpel, Mark W wrote:
>I have noticed that I get exactly the same number of sig. t-tests when
>using the limma function classifyTestsP when using both methods="none"
>and methods="fdr". Delving a little deeper, I performed the function
>p.adjust(fit2$p.value, method="fdr"). Output from this indicates that
>the p values are indeed being properly adjusted with the fdr method.
>
>Why are these adjusted p values not being passed to the classifyTestsP
>function?

Please read the documentation entry for classifyTestsP(), in particular the 
last sentence of the first paragraph under "Details".

I am guessing that you only have one contrast in your linear model, in 
which case classifyTestsP() doesn't do any p-value adjustment.

I know that things may not be obvious, but it is best to make sure that you 
understand the purpose of functions before you announce a "bug" to 
everyone, as in the subject title of your email.

Gordon

>Thanks,
>
>Mark
>
>Mark W. Kimpel MD
>
>Official Business Address:
>
>Department of Psychiatry
>Indiana University School of Medicine
>Biotechnology, Research, & Training Center
>1345 W. 16th Street
>Indianapolis, IN  46202
>
>Preferred Mailing Address:
>
>15032 Hunter Court
>Westfield, IN  46074
>
>(317) 490-5129 Home, Work, & Mobile
>1-(317)-536-2730 FAX