Protocol Online logo
Top : Forum Archives: : Real-Time PCR

QRT-PCR result analysis - relative transcript level? (Feb/19/2004 )


I would like to know how people usually normalize the amount RNA
they use for a quantitative real-time RT-PCR.

I have used two different ways as reported in papers, but
they gave me two totally different results-!!

One method uses the total RNA normalization with cell numbers.

For example, 100 cells produce 100 ng RNA (total=including rRNA and mRNA), I use 10ng for RT-PCR, found 5 copies of mRNA.
Thus, # of copy/cell = (5/10 * 100 ng)/100cells = 0.5

The other methood uses "relative transcript level(RTL)" which
calculates in this way:

RTL = 1000 * 2^(-del Ct) where
del Ct = Ct (target gene) - Ct (constitutive control)??

However, this method doesn't consider the different cell numbers
where the RNA are from.....

I will greatly appreciate your help. !

Thank you.


As far as I understand the number of cells is already calculated into the factor of the constitutive gene.... e.g. if each cell has 10 mRNA copys of a given gene (GAPDH) at every time, then 1E6 detected copys in your assay tell you that there have been 1E5 cells....



biggrin.gif Thank you Mike!!

It sounds to me that, within the same amount of total RNA used, almost same number of costitutive gene's copy exist...?

If one is not interested in knowing the absolute number of transcripts, but rather interested in knowing the relative amount of expression, then
can he use the Ct values from samples directly to calculate the copy number of transcripts?

I mean, can I use 1000*2^(-Ct) instead of 1000*2^(-del Ct)??

Thank you!


Hi there again!

You could use (Ct) instead of (del Ct), if you are sure you are using the same amount of RNA every time...


with the constitutive gene as control you can reduce deviations to a minimum!

There will be errors from the machinery and/the enzymes, not from you - and pippetting errors and so on. Your photometer has a standard deviation in it's measurements, for example. Maybe the harvesting of the cells is not always equally well, or the lysis of the cells for RNA isolation doesn't work exactly the same all the time. And the RT process will be better or worse form time to time, or your Real Time detection can fluctuate too....

All this can be eliminated by using an internal control, i.e. a constituitve gene's mRNA, which is isolated along with your Gene of Interest's (GOI) mRNA, OD260 measured along, reversely transcribed together with the GOI. So when you detect the number of your GOI in your RT cup and measure the numer of transcripts of your Control OUT OF THE SAME CUP, you can either normalise by the number of Control copys of give the ratio copys GOI/copy Contol.... Therefor, most error-sources excluded!