[BioC] Limma: A values in topTable

[Wu, Xiwei]

Dear list,

I did a search in the mail achive, and did not find anyone reporting the
same issue below. 
I was looking at some cDNA array data using Limma 2.0.2 with
R2.1.1patched under Windows2000. I have three different contrasts. When
I looked at the topTables generated from each contrasts, the A (average
intensity) for all the genes are the same across the three contrasts. I
also observed the same thing using another dataset. Is that what is
supposed to happen, or is there anything wrong with my code? Any help
will be highly appreciated.

Xiwei Wu, MD, PhD 
Assistant Research Scientist
Department of Biomedical Informatics
Beckman Research Insitute
City of Hope National Medical Center
Duarte, CA 91010

========================================================================
======================================
> library(limma)
> targets <- readTargets()
> targets
                                       SlideNumber
FileName Cy3 Cy5 
2005-07-06_04J12 HsKGV_17_(Pos_vs_Neg)           1 2005-07-06_04J12
HsKGV_17_(Pos_vs_Neg).gpr   V  C1 
2005-07-06_04J12 HsKGV_18_(36A_vs_Neg)           2 2005-07-06_04J12
HsKGV_18_(36A_vs_Neg).gpr RZ1  C1 
2005-07-06_04J12 HsKGV_19_(36B_vs_Neg)           3 2005-07-06_04J12
HsKGV_19_(36B_vs_Neg).gpr RZ2  C1 
2005-07-06_04J12 HsKGV_20_(Neg_vs_Pos)           4 2005-07-06_04J12
HsKGV_20_(Neg_vs_Pos).gpr  C1   V 
2005-07-06_04J12 HsKGV_22_(Neg_vs_36A)           5 2005-07-06_04J12
HsKGV_22_(Neg_vs_36A).gpr  C1 RZ1 
2005-07-06_04J12 HsKGV_23_(Neg_vs_36B)           6 2005-07-06_04J12
HsKGV_23_(Neg_vs_36B).gpr  C1 RZ2 

> myfun <- function(x) as.numeric(x$Flags > -49.5)
> RG <- read.maimages(targets$FileName, source="genepix", wt.fun=myfun)
> RGb <- backgroundCorrect(RG, method="normexp", offset=50)
> MA <- normalizeWithinArrays(RGb, layout=getLayout(RG$genes))
> MA2 <- normalizeBetweenArrays(MA, method="scale")
> design <- modelMatrix(targets, ref="C1")
Found unique target names:
 C1 RZ1 RZ2 V 
> design
                                       RZ1 RZ2  V
2005-07-06_04J12 HsKGV_17_(Pos_vs_Neg)   0   0 -1
2005-07-06_04J12 HsKGV_18_(36A_vs_Neg)  -1   0  0
2005-07-06_04J12 HsKGV_19_(36B_vs_Neg)   0  -1  0
2005-07-06_04J12 HsKGV_20_(Neg_vs_Pos)   0   0  1
2005-07-06_04J12 HsKGV_22_(Neg_vs_36A)   1   0  0
2005-07-06_04J12 HsKGV_23_(Neg_vs_36B)   0   1  0
> fit <- lmFit(MA2, design)
> cont <- makeContrasts(RZvsV = (RZ1+RZ2)/2-V, VvsC = V, RZvsC =
(RZ1+RZ2)/2, levels=design)
> cont
    RZvsV VvsC RZvsC
RZ1   0.5    0   0.5
RZ2   0.5    0   0.5
V    -1.0    1   0.0
> fit2 <- contrasts.fit(fit, cont)
> fit2 <- eBayes(fit2)
> RZvsV <- topTable(fit2, coef=1, adj="none", n=15360)
> VvsC <- topTable(fit2, coef=2, adj="none", n=15360)
> RZvsC <- topTable(fit2, coef=3, adj="none", n=15360)
> RZvsV_sort <- SortMat(RZvsV, Sort=c(1,2,3))
> VvsC_sort <- SortMat(VvsC, Sort=c(1,2,3))
> RZvsC_sort <- SortMat(RZvsC, Sort=c(1,2,3))

#Below shows that all As are the same
> RZvsV_sort[1:5,]
  Block Row Column  ID                                        Name
M        A         t   P.Value         B
1     1   1      1 Inf                             arylsulfatase B
NA 6.627838        NA        NA        NA
2     1   1      2 Inf                                  epimorphin
NA 6.213439        NA        NA        NA
3     1   1      3 Inf            parathyroid hormone-like hormone
0.11716600 8.375271 0.4991073 0.6311222 -4.696798
4     1   1      4 Inf 5-hydroxytryptamine (serotonin) receptor 2C
NA 6.599125        NA        NA        NA
5     1   1      5 Inf              2,3-bisphosphoglycerate mutase
0.07083349 9.614017 0.2325365 0.8219533 -4.721694
> VvsC_sort[1:5,]
  Block Row Column  ID                                        Name
M        A          t   P.Value         B
1     1   1      1 Inf                             arylsulfatase B
NA 6.627838         NA        NA        NA
2     1   1      2 Inf                                  epimorphin
NA 6.213439         NA        NA        NA
3     1   1      3 Inf            parathyroid hormone-like hormone
-0.03414736 8.375271 -0.1781538 0.8630258 -4.681857
4     1   1      4 Inf 5-hydroxytryptamine (serotonin) receptor 2C
NA 6.599125         NA        NA        NA
5     1   1      5 Inf              2,3-bisphosphoglycerate mutase
0.09565185 9.614017  0.3845843 0.7105558 -4.671349
> RZvsC_sort[1:5,]
  Block Row Column  ID                                        Name
M        A         t   P.Value         B
1     1   1      1 Inf                             arylsulfatase B
NA 6.627838        NA        NA        NA
2     1   1      2 Inf                                  epimorphin
NA 6.213439        NA        NA        NA
3     1   1      3 Inf            parathyroid hormone-like hormone
0.08301864 8.375271 0.6125316 0.5571728 -5.722139
4     1   1      4 Inf 5-hydroxytryptamine (serotonin) receptor 2C
NA 6.599125        NA        NA        NA
5     1   1      5 Inf              2,3-bisphosphoglycerate mutase
0.16648534 9.614017 0.9466494 0.3714985 -5.471472



"EMF <COH.ORG>" made the following annotations.
------------------------------------------------------------------------------
SECURITY/CONFIDENTIALITY WARNING:  This message and any atta...{{dropped}}