[BioC] Twilight

James W. MacDonald jmacdon at med.umich.edu
Thu Feb 21 16:44:23 CET 2008

Hi Mahdi,

Not having used the twilight package before, I took a quick look at the 
help page for both twilight() and twilight.pval() and it seems obvious 
that the functions do two completely different things. Have you looked 
at the help pages? That would seem to be a reasonable place to start. 
See ?twilight and ?twilight.pval

In addition, this package comes with a vignette that discusses both 
functions and what you use them for. Again, this would be a reasonable 
thing to peruse. After loading the package try openVignette() and choose 
the twilight vignette.



Mahdi Osman wrote:
> Dear list,
> I am trying to understand how the package "twilight" works.
> Version:       1.14.1 Title:         Estimation of local false
> discovery rate
> I worked out the example given in the package according to the
> guidelines given. I also tried another run using my own real dataset.
> I am very much interested in getting to understanding a bit more
> about individual outputs of the twilight package. When I execute the
> function "twilight.pval" I get among other things the "result"
> dataframe. I also get the "result" dataframe when I execute the
> function "twilight". I spent a bit of time to figure out if there is
> a difference between the two, ie, "result" from "twilight.pval" and
> "twilight" functions, respectively. I also flipped through the
> guidelines, but could not find a hint on why the "result" dataframe
> is returned in both cases and whether or not there is a difference
> between the two datarames. I would appreciate if anyone can give me
> further hint on when to use "twilight.pval" and "twilight" functions.
> Thanks for your hint in advance.
> Mahdi Osman

James W. MacDonald, M.S.
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109

More information about the Bioconductor mailing list