[BioC] Beginner's question: ebayes

[Eliana Lucchinetti Zaugg]

Dear all:
I've got 30 chips, 5 controls and 5 chips x5 treatment groups and decided that LIMMA would be the best to analyze these data.
I am using this package for the first time (downloaded and installed today, March 16 2004) and I am thus following the manual instructions. But Alas!
I've got the following error message:

fit2 < eBayes(fit2)
  coefficients stdev.unscaled          sigma    df.residual         method         design      contrasts 
          TRUE           TRUE          FALSE           TRUE          FALSE           TRUE          FALSE 
      df.prior       s2.prior      var.prior     proportion        s2.post              t        p.value 
         FALSE          FALSE          FALSE          FALSE           TRUE           TRUE          FALSE 
          lods 
         FALSE 
Warning message: 
longer object length is not a multiple of shorter object length in: fit2 < eBayes(fit2).

Because of this error, no information is extractable afterwards and obviously I get further error messages:
topTable(fit2,coef=1, adjust="fdr")
Error in array(x, c(length(x), 1), if (!is.null(names(x))) list(names(x),  : 
        attempt to set an attribute on NULL
> clas <-classifyTests(fit2)
Error in classifyTests(fit2) : tstat cannot be extracted from object

Any idea of what went wrong?
Thank you very much for any suggestions!
ELIANA

===============================
Eliana Lucchinetti Zaugg, PhD
Institute of Pharmacology and Toxicology 
Section of Cardiovascular Research-Room 17 J 28
University of Zurich
Winterthurerstr. 190
CH-8057 Zürich (Switzerland)
 
Phone: +41-1-635 59 18
Fax: +41-1-635 68 71
e-mail: eliana.zaugg at pharma.unizh.ch


	[[alternative HTML version deleted]]