[BioC] Fwd:decideTests nestedF

[daniela marconi]

Hi, 
I'm using limma to analyze an Operon Oligo data set. 
I would try to use in down stream part of the analysis a nestedF approach. 
The problem is that using 
> results<-decideTest(fit2,method="nestedF",adjust.method="fdr") 
I had this error message 
Errore: nestedF method can't handle NA p-values 

Can I use instead classifyTestsF? What is the difference? 
I will have with this function a way to obtain adjusted p.values 
Or otherwise what is the best solution to avoid NA P.values? 


Below, the R code of my analysis:what is wrong??? 

>Targets<-readTargets() 
>Targets 
file.name Cy5 Cy3 
1 013.gpr nsM linf B 
2 015.gpr UM linf B 
3 018.gpr UM linf B 
4 021.gpr UM linf B 
5 022.gpr UM linf B 
6 032.gpr nsM linf B 
7 039.gpr UM linf B 
8 047.gpr UM linf B 
9 049.gpr nsM linf B 
10 067.gpr sM linf B 
11 068.gpr nsM linf B 
12 079.gpr sM linf B 
13 080.gpr nsM linf B 
14 089.gpr linf B 
15 098.gpr sM linf B 
16 107.gpr sM linf B 
17 119.gpr UM linf B 
18 127.gpr sM linf B 
19 128.gpr UM linf B 
20 129.gpr nsM linf B 
21 149.gpr UM linf B 
22 164.gpr nsM linf B 
23 181.gpr sM linf B 
24 185.gpr sM linf B 
25 186.gpr UM linf B 
26 188.gpr nsM linf B 
27 191.gpr UM linf B 
28 195.gpr UM linf B 
29 245.gpr UM linf B 
30 257.gpr sM linf B 
31 258.gpr nsM linf B 
32 286.gpr nsM linf B 
33 287.gpr sM linf B 
34 288.gpr nsM linf B 
35 304.gpr nsM linf B 
36 305.gpr sM linf B 
37 313.gpr nsM linf B 
38 316.gpr sM linf B 
39 318.gpr sM linf B 
40 320.gpr sM linf B 
41 323.gpr nsM linf B 
42 325.gpr sM linf B 
43 326.gpr nsM linf B 
44 328.gpr UM linf B 
45 329.gpr sM linf B 
46 331.gpr UM linf B 
47 332.gpr sM linf B 
48 334.gpr sM linf B 
49 337.gpr sM linf B 
50 338.gpr sM linf B 
51 340.gpr sM linf B 
52 344.gpr UM linf B 
53 345.gpr UM linf B 
54 346.gpr nsM linf B 
55 354.gpr UM linf B 
56 369.gpr nsM linf B 
57 378.gpr nsM linf B 
58 382.gpr nsM linf B 

>RG<-read.maimages(Targets$file.name,source="genepix",wt.fun=wtflags(0.1)) 
>MAmov<-normalizeWithinArrays(RG,bc.method="movingmin") 

>MAmovQ<-normalizeBetweenArrays(MA,method="quantile") 

>group<-factor(c("nsM",rep("UM",7),"nsM","sM","nsM","sM","nsM","NC",rep("sM",2),"UM","sM","UM","nsM","UM","nsM", 
rep("sM",2),"UM","nsM",rep("UM",3),"sM",rep("nsM",2),"sM",rep("nsM",2),"sM","nsM",rep("sM",3),"nsM","sM","nsM", 
"UM","sM","UM",rep("sM",5),rep("UM",2),"nsM","UM",rep("nsM",3)),levels=c("nsM","UM","sM","NC")) 

>design<-model.matrix(~0+group) 
>colnames(design)<-c("nsM","UM","sM","NC") 
>fit<-lmFit(MAmovQ,design,weights=MAmovQ$weights,ndups=1) 

>cont.matrix<-makeContrasts(UM-(nsM+sM),UM-nsM,UM-sM,nsM-sM,levels=design) 

> cont.matrix
    UM - (nsM + sM) UM - nsM UM - sM nsM - sM
nsM              -1       -1       0        1
UM                1        1       1        0
sM               -1        0      -1       -1
NC                0        0       0        0

>fit2<-contrasts.fit(fit,cont.matrix) 

>results<-decideTests(fit2,method="nestedF",adjust.method="fdr",p.value=0.05)