[BioC] RankProd Error Message

[fhong@salk.edu]

Since there are two classed in your data, the class label(eset.cl) should
have two categories only. Try to use
eset.cl <- rep(c(rep(0,4),rep(1,4)),2)
> eset.cl
 [1] 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1

then your class "INF" would be class 1 and "NONINF" would be class 2. And
make sure your eset_rma is not a affybatch object, but a matrix-format
data set.

But if your experiment is pairwise design,meaning 1 and -1 is a pair, 2
and -2 is a pair and so on, then you need to get ratio of 1/-1, 2/-2,...
and use the "one-class" option in function RP.

Let me know what you get.

Bests,
Fangxin



> Hello all!
>
> I'm trying to analyze 16 Affy arrays with SAM, Limma and RankProd to
> look at differential expression.
>
> I got the following error message from the RP function.  Any help as to
> what it means?
>
>> RP.out<- RP(eset_rma, eset.cl, num.perm = 100, logged=TRUE,
> na.rm=FALSE, plot=FALSE, gene.names=eset.gnames)
> Error in if (total.sam != total.sam2) stop("Number of classes should
> match the columns in the data") :
>         argument is of length zero
>
> I have included below the phenotype data of eset_rma and the listing of
> eset.cl:
>
>> pData(eset_rma)
>                    SAMPLE_NUM TREATMENT TCELLS
> NE_MDM-INF1.CEL             1       INF      N
> NE_MDM-INF2.CEL             2       INF      N
> NE_MDM-INF3.CEL             3       INF      N
> NE_MDM-INF4.CEL             4       INF      N
> NE_MDM-T-1.CEL              5    NONINF      Y
> NE_MDM-T-2.CEL              6    NONINF      Y
> NE_MDM-T-3.CEL              7    NONINF      Y
> NE_MDM-T-4.CEL              8    NONINF      Y
> NE_MDM-T-INF-1.CEL          9       INF      Y
> NE_MDM-T-INF-2.CEL         10       INF      Y
> NE_MDM-T-INF-3.CEL         11       INF      Y
> NE_MDM-T-INF-4.CEL         12       INF      Y
> NE_MDM1.CEL                13    NONINF      N
> NE_MDM2.CEL                14    NONINF      N
> NE_MDM3.CEL                15    NONINF      N
> NE_MDM4.CEL                16    NONINF      N
>
>> eset.cl
>  [1]  1  2  3  4 -5 -6 -7 -8  5  6  7  8 -1 -2 -3 -4
>
> Thanks for any help you can provide!
>
> ---Nick Ettinger
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
>
>


--------------------
Fangxin Hong  Ph.D.
Plant Biology Laboratory
The Salk Institute
10010 N. Torrey Pines Rd.
La Jolla, CA 92037
E-mail: fhong at salk.edu
(Phone): 858-453-4100 ext 1105