[BioC] HTqPCR problem with replicates

[Marco Fabbri]

Hi,

I am using TLDA cards with HTqPCR, the cards have 2 samples for each
card and the features are in triplicates. I import the card and run
limma but I realised that the triplicates are kept apart. As you can
see from the last three lines 18S is presented three times but I would
like to have a average of the three replicates.

qPCRraw= readCtData(files=c("palte_res.txt"), path = NULL, n.features
= 384, flag = 4, feature = 6, type = 7, position = 3, Ct = 8, header =
TRUE, SDS = TRUE,  na.value = 40,n.data=2)

###set the layout (192 for each card)
sample2a.order <- factor(rep(c("x","Y"), each=192))
qPCRnew <- changeCtLayout(qPCRraw[,], sample.order = sample2a.order)
###change the name
sampleNames(qPCRnew) <- c("Ctrl_R1","Diuron50_R1","Ctrl_R2","Diuron50_R2")
#set the group
groups = factor(c("Ctrl","Diuron50","Ctrl","Diuron50"))

q2.cat=setCategory(qPCRnew, Ct.max = 35,groups = samples,n.category=1)
q2.cat=filterCtData(qPCRnew,remove.category="Undetermined" ,n.category=1)
#normalized
q2.NORM= normalizeCtData(
q2.cat,norm="deltaCt",deltaCt.genes=c("ABL1-Hs01104728_m1","MRPL19-Hs01040217_m1","HPRT1-Hs03929098_m1"))

######limma

design <- model.matrix(~0 + groups)
colnames(design) <- c("Ctrl","Diuron50")
print(design)

contrasts <- makeContrasts(Ctrl-Diuron50, levels = design)

qDE.limma <- limmaCtData(q2.NORM, design = design, contrasts = contrasts)

 qDE.limma
$`Ctrl - Diuron50`
                            genes feature.pos       t.test
p.value adj.p.value         ddCt         FC   meanTarget
meanCalibrator
10              18S-Hs99999901_s1        <NA> 13.020948738
0.0006195627  0.04484078  2.856187444 0.13810262 -12.10495367
-14.961141111
11              18S-Hs99999901_s1        <NA> 11.461949841
0.0009341829  0.04484078  2.958459694 0.12865151 -12.17514617
-15.133605861
12              18S-Hs99999901_s1        <NA> 11.851588812
0.0008389948  0.04484078  2.681318944 0.15589873 -12.30213467
-14.98345361