[BioC] HTqPCR - TLDA analysis doubt

[Heidi Dvinge]

Hi Deepak,

> Iam using HTqPCR for my TLDA analysis and i have some doubt in that
> 1)	In Fold change or relative quantification analysis plots two
> legends are their target undetermined and calibrator undetermined
> could please tell me what is meant by that.

If you look at the example in ?plotCtRQ, you'll see this:

# Load example data and calculate differential expression
data(qPCRpros)
qDE <- ttestCtData(qPCRpros[,1:4], groups=factor(c("A", "B", "B", "A")),
calibrator="B")
# Plotting the top 10 results or first 10 genes
plotCtRQ(qDE, genes=1:10)

This means that the two samples groups called "A" and "B" are compared,
with "A" in this case being considered the target sample and "B" the
calibrator. More information is given in the qDE object:

head(qDE)

In this case, you do get that there's a significant difference in Ct value
between group A and B. However, this might be affected by how reliable the
original Ct values were. The boxes are therefore marked with shading
lines, to make you aware that these results should perhaps be taken with a
grain of salt, and that's it's worth going back to the original data and
checking how many of the Ct values were reliable.

In you case you might have that all the values are "OK", i.e. none are
unreliable, hence none of the boxes are marked. As explained in ?plotCtRQ
you can remove the shading lines by setting mark.un=FALSE, such as

plotCtRQ(qDE, genes=1:10, mark.un=FALSE)


> 2)	During classifying the genes bases on the function it is showing
> some errors. I have 47 gene for one sample.
> Is there any error in the following
>>featureClass(raw) <- factor(c("Onc", "Apop", "Cyclin", " Cyclin", "Onc",
>> "Cyc kinase", "cyc inhibitor", "TS", "Onc","TS","Onc",
>> TS")[sample(c(2,3,16,20,26,33,36,38,39,40,47), 47, replace = TRUE)])
> +
>
>> featureClass(raw) <- factor(rep(c("Anti_Apoptotic",
>> "Cell_Cycle_Inhibitor", "Cell_Proliferators", " Cyclin",
>> "Cyclin_Dependent_Kinase", "Oncogene", "Protein_Degradation",
>> "Tumor_Supressor"), times=c(1,11,4,13,6,3,3,6)))
>> plotCtCategory(raw.cat, stratify = "class")
> Error in split.default(data[, i], strat) :
>   Group length is 0 but data length > 0
>> plotCtCategory(q.norm,stratify = "class")
> Error in split.default(data[, i], strat) :
>   Group length is 0 but data length > 0
>
I'm afraid I can't reproduce this with my data. For example, if I use 47
genes and 3 samples I can say:

> data(qPCRraw)
> raw	<- qPCRraw[1:47,1]
> raw
An object of class "qPCRset"
Size:  47 features, 1 samples
Feature types:		 Endogenous Control, Target
Feature names:		 Gene1 Gene2 Gene3 ...
Feature classes:	 Kinase, Marker, TF
Feature categories:	 OK, Undetermined
Sample names:		 sample1 NA NA ...
> featureClass(raw) <- factor(rep(c("Anti_Apoptotic",
"Cell_Cycle_Inhibitor", "Cell_Proliferators", " Cyclin",
"Cyclin_Dependent_Kinase", "Oncogene","Protein_Degradation",
"Tumor_Supressor"), times=c(1,11,4,13,6,3,3,6)))
> par(mar=c(10,2,2,2))
> plotCtCategory(raw, stratify = "class", xlim=c(0,9), ylim=c(0,16))

Can you perhaps provide me with more information about you data? Please
also note than in the code you provide, you set featureClass(RAW) <- ...
but try to plot plotCtCategory(RAW.CAT, stratify = "class"), i.e. a
different object. Does raw.cat have any feature classes assigned ot it?

HTH
\Heidi


>
>> sessionInfo()
> R version 2.11.0 (2010-04-22)
> x86_64-pc-mingw32
>
> locale:
> [1] LC_COLLATE=English_India.1252  LC_CTYPE=English_India.1252
> [3] LC_MONETARY=English_India.1252 LC_NUMERIC=C
> [5] LC_TIME=English_India.1252
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
> other attached packages:
> [1] HTqPCR_1.2.0       limma_3.4.3        RColorBrewer_1.0-2 Biobase_2.8.0
>
> loaded via a namespace (and not attached):
> [1] affy_1.26.1           affyio_1.16.0         gdata_2.8.0
> [4] gplots_2.8.0          gtools_2.6.2          preprocessCore_1.10.0
>>
>
> --
>