Normalization of Ct of interest to a reference Ct - (Apr/29/2013 )
I face some difficulties with the normalized Ct values.. I do not have a calibrator, I just observe differences, thus, I just need a way to
normalize my data and compare them with the endogenous control simultaneously .
I have read somewhere that one way for normalizing PCR data is to divide the Ct of the gene of interest by the Ct obtained from a
housekeeping/reference gene (Ct inter. / Ct refer.). However, I am not sure how to interpret the resulting values. Is it true that the larger
the ratio, the lower the expression of the gene of interest?
Additionaly, what about the 2-ΔCt (ΔCt = Ct inter. - Ct refer.) ? What does thi s value depict? Provided that 2-ΔCt = X, can I state something
like: "The expression was observed to be X copies of the gene of interest mRNA / copy of reference gene mRNA" ? Finally, 2-ΔCt values
less than one means that the expession of the gene of interest is less than the expression of the reference one, whereas for 2-ΔCt values
more than one holds the opposite?
Any help is more than welcome !!!!!!
Both methods of normalization are fairly common and they both tell similar stories: what's the difference between two genes, considering that they have this much reference gene expression. The only thing that's important is that for a given question, you have to always use the same method.
In the first case, in general, the larger the ratio the lower the expression of your gene of interest. In qPCR, fewer cycles means fewer time to hit the threshold cycle and therefore more initial product. If you have a really large numerator (high cycles) and a really small denominator (low cycles), then you can read that independently as less starting material for the target of interest and more starting material for the reference. This will give you a larger ratio. If you had a smaller numerator, then you had more starting material, and therefore a smaller ratio.
The same goes for the second case.
The values simply depict the the relative levels of one gene compared to another.
Without a parallel run standard curve of pre-quantified material you cannot make any claims to copy number of a gene. You can only report relative levels and fold changes.
If your subtracted normalized values come out to less than one it means that your target of interest hit the detection threshold quicker than your reference gene. That generally means that you had more starting material for the target of interest, but that also depends on the size of that amplicon versus the size of the reference amplicon. All things being equal, you had more of your target than of your reference. If your reference is tiny and your target is huge, then you have to reconsider the whole story.
I hope this helps.
Thank you JYaron!
is any of you experienced standard curve for relative quantification? I want to measure relative changes in breast tumor tissue. I want to ask what samples could I use to generate it? Can I use any of breast tumor tissue samples to generate the standard curve?