[BioC] RE : Re: RE : Re: siggesn and ebam

[Holger Schwender]

Yes, in ebam missing values are not allowed. I was also thinking about removing the missing value handling stuff from the sam functions such that also in sam missing values are not allowed, but I didn't do this. Please note that also SAM cannot handle missing values. So genes with missing values are either removed or missing values are imputed in sam before the actual SAM analysis starts.

Without knowing the error message it is hard to say something about your second question. But I think that this is not an error message of sam but of data.frame, as each row in a data frame must have another name. So slightly rename your replicates (e.g., use not twice geneXXX but once geneXXXa and once geneXXXb as the name for this particular gene) and it should work.

Holger


-------- Original-Nachricht --------
Datum: Fri, 20 Jul 2007 11:54:13 +0200 (CEST)
Von: el mousselly antra <elmousselly.antra at yahoo.fr>
An: Holger Schwender <holger.schw at gmx.de>
CC: bio c bioconductor <bioconductor at stat.math.ethz.ch>
Betreff: RE : Re: RE : Re: siggesn and ebam

> hi
> holger ,
> i have done ebam i think that this not tolerat   missing data like
> function 
> sam and another think 
> when we have a repetition of spots on the same array 
> sam function  have a message error for some functions 
>   ???
> thanks 
> best wishes
>   elmousselly antra
> 
> 
> Holger Schwender <holger.schw at gmx.de> a écrit :
>   If ebam.out is the output of ebam und you want to use delta=0.9 as the
> minimum posterior probability for your genes to be called differentially
> expressed, then
> 
> > summary(ebam.out, delta = 0.9)
> 
> returns the genes called differentially expressed with other useful
> statistics as their z-value and their local FDR, but no raw p-value is returned
> as in sam.
> 
> If you are just interested in the names of this genes you can also use
> 
> > list.siggenes(ebam.out, 0.9)
> 
> Best,
> Holger
> 
> -------- Original-Nachricht --------
> Datum: Fri, 20 Jul 2007 09:04:38 +0200 (CEST)
> Von: el mousselly antra 
> An: Holger Schwender 
> CC: bio c bioconductor 
> Betreff: RE : Re: sam and the new version
> 
> > 
> > hi holger,
> > 1.yes siggenes work well, the problem was in the 
> > version i use an old version.there is no problems now
> > 2.but i have an other problem to have a list of genes DE on 
> > ebam and also p-value for each genes i don't know how 
> > because in the vignette i don't see something that explain this 
> > you talk just about number of genes DE but 
> > list of genes DE ???
> > thanks 
> > best wishes
> > elmousselly antra 
> > 
> > Holger Schwender a écrit :
> > Actually, I do not know anymore what works in 1.2.0 and not in 1.2.11,
> > as these are pretty old versions which have been updated several times
> with
> > many major changes. Please use the current version of siggenes, version
> > 1.10.1.
> > 
> > And please let me then know if something does not work in this version.
> > And if yes, what does not work.
> > 
> > If you do not know how to specify the arguments/options in the function,
> > please take a look in the vignette of siggenes:
> > 
> > > vignette("siggenes")
> > 
> > Almost everything is explained in this vignette (and what is not
> explained
> > in the vignette is explained in the R News article mentioned -- with
> link
> > -- in the vignette).
> > 
> > Thanks,
> > Holger
> > 
> > -------- Original-Nachricht --------
> > Datum: Thu, 19 Jul 2007 09:17:34 +0200 (CEST)
> > Von: el mousselly antra 
> > An: holger.schw at gmx.de
> > CC: bio c bioconductor 
> > Betreff: sam and the new version
> > 
> > > dear schwender,
> > > i am using sam in siggens but i have some problems
> > > some functions don't work as i have the habit to use it, like
> > > summary(sam.out)
> > > sum.sam.out <- summary(sam.out, 0.4, ll = FALSE)
> > > sum.sam.out at mat.sig
> > > only function sam work well but a lot of functions functions 
> > > have some problems or changes that i don'know how but i think
> > > in the version 1.2.0 in the past 
> > > have worked but now this new version d'ont work in the version 1.2.11
> > > i don't know what'the problem and what's i have to do
> > > thank you 
> > > sincerly
> > > 
> > > 
> > > ---------------------------------
> > > Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers
> Yahoo!
> > > Mail 
> > 
> > -- 
> > Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten 
> > Browser-Versionen downloaden: http://www.gmx.net/de/go/browser
> > 
> > -- 
> > Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten 
> > Browser-Versionen downloaden: http://www.gmx.net/de/go/browser
> > 
> > 
> > 
> > ---------------------------------
> > Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo!
> > Mail 
> 
> -- 
> Psssst! Schon vom neuen GMX MultiMessenger gehört?
> Der kanns mit allen: http://www.gmx.net/de/go/multimessenger
> 
> 
>        
> ---------------------------------
>  Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo!
> Mail 

-- 
Psssst! Schon vom neuen GMX MultiMessenger gehört?
Der kanns mit allen: http://www.gmx.net/de/go/multimessenger